2.8

- 新增 RequestCoordinator，请求统一走限流、冷却、排队与派发流程
- 新增 token 估算与 UsageTracker，并接入 API usage 解析（prompt/completion/total）
- 新增上下文 token 预算裁剪与可选自动总结能力
- 重构命令调用链至协调器，新增 `/chatgpt stats [reset]`，同步补全权限与补全提示
- 新增 ConfigFileUpdater：为 `config.yml` 与 `lang/*.yml` 自动补齐缺失项，保留用户已有配置与注释，并生成备份
- 修复自动补全插入顺序/空行/注释归属问题，避免重复块与空行膨胀
- 优化 `config.yml` 注释说明：保持原意，仅补充更清晰的中英文细节说明

LICENSE

@@ -0,0 +1,661 @@
+                    GNU AFFERO GENERAL PUBLIC LICENSE
+                       Version 3, 19 November 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+                            Preamble
+
+  The GNU Affero General Public License is a free, copyleft license for
+software and other kinds of works, specifically designed to ensure
+cooperation with the community in the case of network server software.
+
+  The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works.  By contrast,
+our General Public Licenses are intended to guarantee your freedom to
+share and change all versions of a program--to make sure it remains free
+software for all its users.
+
+  When we speak of free software, we are referring to freedom, not
+price.  Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+them if you wish), that you receive source code or can get it if you
+want it, that you can change the software or use pieces of it in new
+free programs, and that you know you can do these things.
+
+  Developers that use our General Public Licenses protect your rights
+with two steps: (1) assert copyright on the software, and (2) offer
+you this License which gives you legal permission to copy, distribute
+and/or modify the software.
+
+  A secondary benefit of defending all users' freedom is that
+improvements made in alternate versions of the program, if they
+receive widespread use, become available for other developers to
+incorporate.  Many developers of free software are heartened and
+encouraged by the resulting cooperation.  However, in the case of
+software used on network servers, this result may fail to come about.
+The GNU General Public License permits making a modified version and
+letting the public access it on a server without ever releasing its
+source code to the public.
+
+  The GNU Affero General Public License is designed specifically to
+ensure that, in such cases, the modified source code becomes available
+to the community.  It requires the operator of a network server to
+provide the source code of the modified version running there to the
+users of that server.  Therefore, public use of a modified version, on
+a publicly accessible server, gives the public access to the source
+code of the modified version.
+
+  An older license, called the Affero General Public License and
+published by Affero, was designed to accomplish similar goals.  This is
+a different license, not a version of the Affero GPL, but Affero has
+released a new version of the Affero GPL which permits relicensing under
+this license.
+
+  The precise terms and conditions for copying, distribution and
+modification follow.
+
+                       TERMS AND CONDITIONS
+
+  0. Definitions.
+
+  "This License" refers to version 3 of the GNU Affero General Public License.
+
+  "Copyright" also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+  "The Program" refers to any copyrightable work licensed under this
+License.  Each licensee is addressed as "you".  "Licensees" and
+"recipients" may be individuals or organizations.
+
+  To "modify" a work means to copy from or adapt all or part of the work
+in a fashion requiring copyright permission, other than the making of an
+exact copy.  The resulting work is called a "modified version" of the
+earlier work or a work "based on" the earlier work.
+
+  A "covered work" means either the unmodified Program or a work based
+on the Program.
+
+  To "propagate" a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for
+infringement under applicable copyright law, except executing it on a
+computer or modifying a private copy.  Propagation includes copying,
+distribution (with or without modification), making available to the
+public, and in some countries other activities as well.
+
+  To "convey" a work means any kind of propagation that enables other
+parties to make or receive copies.  Mere interaction with a user through
+a computer network, with no transfer of a copy, is not conveying.
+
+  An interactive user interface displays "Appropriate Legal Notices"
+to the extent that it includes a convenient and prominently visible
+feature that (1) displays an appropriate copyright notice, and (2)
+tells the user that there is no warranty for the work (except to the
+extent that warranties are provided), that licensees may convey the
+work under this License, and how to view a copy of this License.  If
+the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+  1. Source Code.
+
+  The "source code" for a work means the preferred form of the work
+for making modifications to it.  "Object code" means any non-source
+form of a work.
+
+  A "Standard Interface" means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of
+interfaces specified for a particular programming language, one that
+is widely used among developers working in that language.
+
+  The "System Libraries" of an executable work include anything, other
+than the work as a whole, that (a) is included in the normal form of
+packaging a Major Component, but which is not part of that Major
+Component, and (b) serves only to enable use of the work with that
+Major Component, or to implement a Standard Interface for which an
+implementation is available to the public in source code form.  A
+"Major Component", in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system
+(if any) on which the executable work runs, or a compiler used to
+produce the work, or an object code interpreter used to run it.
+
+  The "Corresponding Source" for a work in object code form means all
+the source code needed to generate, install, and (for an executable
+work) run the object code and to modify the work, including scripts to
+control those activities.  However, it does not include the work's
+System Libraries, or general-purpose tools or generally available free
+programs which are used unmodified in performing those activities but
+which are not part of the work.  For example, Corresponding Source
+includes interface definition files associated with source files for
+the work, and the source code for shared libraries and dynamically
+linked subprograms that the work is specifically designed to require,
+such as by intimate data communication or control flow between those
+subprograms and other parts of the work.
+
+  The Corresponding Source need not include anything that users
+can regenerate automatically from other parts of the Corresponding
+Source.
+
+  The Corresponding Source for a work in source code form is that
+same work.
+
+  2. Basic Permissions.
+
+  All rights granted under this License are granted for the term of
+copyright on the Program, and are irrevocable provided the stated
+conditions are met.  This License explicitly affirms your unlimited
+permission to run the unmodified Program.  The output from running a
+covered work is covered by this License only if the output, given its
+content, constitutes a covered work.  This License acknowledges your
+rights of fair use or other equivalent, as provided by copyright law.
+
+  You may make, run and propagate covered works that you do not
+convey, without conditions so long as your license otherwise remains
+in force.  You may convey covered works to others for the sole purpose
+of having them make modifications exclusively for you, or provide you
+with facilities for running those works, provided that you comply with
+the terms of this License in conveying all material for which you do
+not control copyright.  Those thus making or running the covered works
+for you must do so exclusively on your behalf, under your direction
+and control, on terms that prohibit them from making any copies of
+your copyrighted material outside their relationship with you.
+
+  Conveying under any other circumstances is permitted solely under
+the conditions stated below.  Sublicensing is not allowed; section 10
+makes it unnecessary.
+
+  3. Protecting Users' Legal Rights From Anti-Circumvention Law.
+
+  No covered work shall be deemed part of an effective technological
+measure under any applicable law fulfilling obligations under article
+11 of the WIPO copyright treaty adopted on 20 December 1996, or
+similar laws prohibiting or restricting circumvention of such
+measures.
+
+  When you convey a covered work, you waive any legal power to forbid
+circumvention of technological measures to the extent such circumvention
+is effected by exercising rights under this License with respect to
+the covered work, and you disclaim any intention to limit operation or
+modification of the work as a means of enforcing, against the work's
+users, your or third parties' legal rights to forbid circumvention of
+technological measures.
+
+  4. Conveying Verbatim Copies.
+
+  You may convey verbatim copies of the Program's source code as you
+receive it, in any medium, provided that you conspicuously and
+appropriately publish on each copy an appropriate copyright notice;
+keep intact all notices stating that this License and any
+non-permissive terms added in accord with section 7 apply to the code;
+keep intact all notices of the absence of any warranty; and give all
+recipients a copy of this License along with the Program.
+
+  You may charge any price or no price for each copy that you convey,
+and you may offer support or warranty protection for a fee.
+
+  5. Conveying Modified Source Versions.
+
+  You may convey a work based on the Program, or the modifications to
+produce it from the Program, in the form of source code under the
+terms of section 4, provided that you also meet all of these conditions:
+
+    a) The work must carry prominent notices stating that you modified
+    it, and giving a relevant date.
+
+    b) The work must carry prominent notices stating that it is
+    released under this License and any conditions added under section
+    7.  This requirement modifies the requirement in section 4 to
+    "keep intact all notices".
+
+    c) You must license the entire work, as a whole, under this
+    License to anyone who comes into possession of a copy.  This
+    License will therefore apply, along with any applicable section 7
+    additional terms, to the whole of the work, and all its parts,
+    regardless of how they are packaged.  This License gives no
+    permission to license the work in any other way, but it does not
+    invalidate such permission if you have separately received it.
+
+    d) If the work has interactive user interfaces, each must display
+    Appropriate Legal Notices; however, if the Program has interactive
+    interfaces that do not display Appropriate Legal Notices, your
+    work need not make them do so.
+
+  A compilation of a covered work with other separate and independent
+works, which are not by their nature extensions of the covered work,
+and which are not combined with it such as to form a larger program,
+in or on a volume of a storage or distribution medium, is called an
+"aggregate" if the compilation and its resulting copyright are not
+used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit.  Inclusion of a covered work
+in an aggregate does not cause this License to apply to the other
+parts of the aggregate.
+
+  6. Conveying Non-Source Forms.
+
+  You may convey a covered work in object code form under the terms
+of sections 4 and 5, provided that you also convey the
+machine-readable Corresponding Source under the terms of this License,
+in one of these ways:
+
+    a) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by the
+    Corresponding Source fixed on a durable physical medium
+    customarily used for software interchange.
+
+    b) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by a
+    written offer, valid for at least three years and valid for as
+    long as you offer spare parts or customer support for that product
+    model, to give anyone who possesses the object code either (1) a
+    copy of the Corresponding Source for all the software in the
+    product that is covered by this License, on a durable physical
+    medium customarily used for software interchange, for a price no
+    more than your reasonable cost of physically performing this
+    conveying of source, or (2) access to copy the
+    Corresponding Source from a network server at no charge.
+
+    c) Convey individual copies of the object code with a copy of the
+    written offer to provide the Corresponding Source.  This
+    alternative is allowed only occasionally and noncommercially, and
+    only if you received the object code with such an offer, in accord
+    with subsection 6b.
+
+    d) Convey the object code by offering access from a designated
+    place (gratis or for a charge), and offer equivalent access to the
+    Corresponding Source in the same way through the same place at no
+    further charge.  You need not require recipients to copy the
+    Corresponding Source along with the object code.  If the place to
+    copy the object code is a network server, the Corresponding Source
+    may be on a different server (operated by you or a third party)
+    that supports equivalent copying facilities, provided you maintain
+    clear directions next to the object code saying where to find the
+    Corresponding Source.  Regardless of what server hosts the
+    Corresponding Source, you remain obligated to ensure that it is
+    available for as long as needed to satisfy these requirements.
+
+    e) Convey the object code using peer-to-peer transmission, provided
+    you inform other peers where the object code and Corresponding
+    Source of the work are being offered to the general public at no
+    charge under subsection 6d.
+
+  A separable portion of the object code, whose source code is excluded
+from the Corresponding Source as a System Library, need not be
+included in conveying the object code work.
+
+  A "User Product" is either (1) a "consumer product", which means any
+tangible personal property which is normally used for personal, family,
+or household purposes, or (2) anything designed or sold for incorporation
+into a dwelling.  In determining whether a product is a consumer product,
+doubtful cases shall be resolved in favor of coverage.  For a particular
+product received by a particular user, "normally used" refers to a
+typical or common use of that class of product, regardless of the status
+of the particular user or of the way in which the particular user
+actually uses, or expects or is expected to use, the product.  A product
+is a consumer product regardless of whether the product has substantial
+commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+  "Installation Information" for a User Product means any methods,
+procedures, authorization keys, or other information required to install
+and execute modified versions of a covered work in that User Product from
+a modified version of its Corresponding Source.  The information must
+suffice to ensure that the continued functioning of the modified object
+code is in no case prevented or interfered with solely because
+modification has been made.
+
+  If you convey an object code work under this section in, or with, or
+specifically for use in, a User Product, and the conveying occurs as
+part of a transaction in which the right of possession and use of the
+User Product is transferred to the recipient in perpetuity or for a
+fixed term (regardless of how the transaction is characterized), the
+Corresponding Source conveyed under this section must be accompanied
+by the Installation Information.  But this requirement does not apply
+if neither you nor any third party retains the ability to install
+modified object code on the User Product (for example, the work has
+been installed in ROM).
+
+  The requirement to provide Installation Information does not include a
+requirement to continue to provide support service, warranty, or updates
+for a work that has been modified or installed by the recipient, or for
+the User Product in which it has been modified or installed.  Access to a
+network may be denied when the modification itself materially and
+adversely affects the operation of the network or violates the rules and
+protocols for communication across the network.
+
+  Corresponding Source conveyed, and Installation Information provided,
+in accord with this section must be in a format that is publicly
+documented (and with an implementation available to the public in
+source code form), and must require no special password or key for
+unpacking, reading or copying.
+
+  7. Additional Terms.
+
+  "Additional permissions" are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions.
+Additional permissions that are applicable to the entire Program shall
+be treated as though they were included in this License, to the extent
+that they are valid under applicable law.  If additional permissions
+apply only to part of the Program, that part may be used separately
+under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+  When you convey a copy of a covered work, you may at your option
+remove any additional permissions from that copy, or from any part of
+it.  (Additional permissions may be written to require their own
+removal in certain cases when you modify the work.)  You may place
+additional permissions on material, added by you to a covered work,
+for which you have or can give appropriate copyright permission.
+
+  Notwithstanding any other provision of this License, for material you
+add to a covered work, you may (if authorized by the copyright holders of
+that material) supplement the terms of this License with terms:
+
+    a) Disclaiming warranty or limiting liability differently from the
+    terms of sections 15 and 16 of this License; or
+
+    b) Requiring preservation of specified reasonable legal notices or
+    author attributions in that material or in the Appropriate Legal
+    Notices displayed by works containing it; or
+
+    c) Prohibiting misrepresentation of the origin of that material, or
+    requiring that modified versions of such material be marked in
+    reasonable ways as different from the original version; or
+
+    d) Limiting the use for publicity purposes of names of licensors or
+    authors of the material; or
+
+    e) Declining to grant rights under trademark law for use of some
+    trade names, trademarks, or service marks; or
+
+    f) Requiring indemnification of licensors and authors of that
+    material by anyone who conveys the material (or modified versions of
+    it) with contractual assumptions of liability to the recipient, for
+    any liability that these contractual assumptions directly impose on
+    those licensors and authors.
+
+  All other non-permissive additional terms are considered "further
+restrictions" within the meaning of section 10.  If the Program as you
+received it, or any part of it, contains a notice stating that it is
+governed by this License along with a term that is a further
+restriction, you may remove that term.  If a license document contains
+a further restriction but permits relicensing or conveying under this
+License, you may add to a covered work material governed by the terms
+of that license document, provided that the further restriction does
+not survive such relicensing or conveying.
+
+  If you add terms to a covered work in accord with this section, you
+must place, in the relevant source files, a statement of the
+additional terms that apply to those files, or a notice indicating
+where to find the applicable terms.
+
+  Additional terms, permissive or non-permissive, may be stated in the
+form of a separately written license, or stated as exceptions;
+the above requirements apply either way.
+
+  8. Termination.
+
+  You may not propagate or modify a covered work except as expressly
+provided under this License.  Any attempt otherwise to propagate or
+modify it is void, and will automatically terminate your rights under
+this License (including any patent licenses granted under the third
+paragraph of section 11).
+
+  However, if you cease all violation of this License, then your
+license from a particular copyright holder is reinstated (a)
+provisionally, unless and until the copyright holder explicitly and
+finally terminates your license, and (b) permanently, if the copyright
+holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+  Moreover, your license from a particular copyright holder is
+reinstated permanently if the copyright holder notifies you of the
+violation by some reasonable means, this is the first time you have
+received notice of violation of this License (for any work) from that
+copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+  Termination of your rights under this section does not terminate the
+licenses of parties who have received copies or rights from you under
+this License.  If your rights have been terminated and not permanently
+reinstated, you do not qualify to receive new licenses for the same
+material under section 10.
+
+  9. Acceptance Not Required for Having Copies.
+
+  You are not required to accept this License in order to receive or
+run a copy of the Program.  Ancillary propagation of a covered work
+occurring solely as a consequence of using peer-to-peer transmission
+to receive a copy likewise does not require acceptance.  However,
+nothing other than this License grants you permission to propagate or
+modify any covered work.  These actions infringe copyright if you do
+not accept this License.  Therefore, by modifying or propagating a
+covered work, you indicate your acceptance of this License to do so.
+
+  10. Automatic Licensing of Downstream Recipients.
+
+  Each time you convey a covered work, the recipient automatically
+receives a license from the original licensors, to run, modify and
+propagate that work, subject to this License.  You are not responsible
+for enforcing compliance by third parties with this License.
+
+  An "entity transaction" is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an
+organization, or merging organizations.  If propagation of a covered
+work results from an entity transaction, each party to that
+transaction who receives a copy of the work also receives whatever
+licenses to the work the party's predecessor in interest had or could
+give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if
+the predecessor has it or can get it with reasonable efforts.
+
+  You may not impose any further restrictions on the exercise of the
+rights granted or affirmed under this License.  For example, you may
+not impose a license fee, royalty, or other charge for exercise of
+rights granted under this License, and you may not initiate litigation
+(including a cross-claim or counterclaim in a lawsuit) alleging that
+any patent claim is infringed by making, using, selling, offering for
+sale, or importing the Program or any portion of it.
+
+  11. Patents.
+
+  A "contributor" is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based.  The
+work thus licensed is called the contributor's "contributor version".
+
+  A contributor's "essential patent claims" are all patent claims
+owned or controlled by the contributor, whether already acquired or
+hereafter acquired, that would be infringed by some manner, permitted
+by this License, of making, using, or selling its contributor version,
+but do not include claims that would be infringed only as a
+consequence of further modification of the contributor version.  For
+purposes of this definition, "control" includes the right to grant
+patent sublicenses in a manner consistent with the requirements of
+this License.
+
+  Each contributor grants you a non-exclusive, worldwide, royalty-free
+patent license under the contributor's essential patent claims, to
+make, use, sell, offer for sale, import and otherwise run, modify and
+propagate the contents of its contributor version.
+
+  In the following three paragraphs, a "patent license" is any express
+agreement or commitment, however denominated, not to enforce a patent
+(such as an express permission to practice a patent or covenant not to
+sue for patent infringement).  To "grant" such a patent license to a
+party means to make such an agreement or commitment not to enforce a
+patent against the party.
+
+  If you convey a covered work, knowingly relying on a patent license,
+and the Corresponding Source of the work is not available for anyone
+to copy, free of charge and under the terms of this License, through a
+publicly available network server or other readily accessible means,
+then you must either (1) cause the Corresponding Source to be so
+available, or (2) arrange to deprive yourself of the benefit of the
+patent license for this particular work, or (3) arrange, in a manner
+consistent with the requirements of this License, to extend the patent
+license to downstream recipients.  "Knowingly relying" means you have
+actual knowledge that, but for the patent license, your conveying the
+covered work in a country, or your recipient's use of the covered work
+in a country, would infringe one or more identifiable patents in that
+country that you have reason to believe are valid.
+
+  If, pursuant to or in connection with a single transaction or
+arrangement, you convey, or propagate by procuring conveyance of, a
+covered work, and grant a patent license to some of the parties
+receiving the covered work authorizing them to use, propagate, modify
+or convey a specific copy of the covered work, then the patent license
+you grant is automatically extended to all recipients of the covered
+work and works based on it.
+
+  A patent license is "discriminatory" if it does not include within
+the scope of its coverage, prohibits the exercise of, or is
+conditioned on the non-exercise of one or more of the rights that are
+specifically granted under this License.  You may not convey a covered
+work if you are a party to an arrangement with a third party that is
+in the business of distributing software, under which you make payment
+to the third party based on the extent of your activity of conveying
+the work, and under which the third party grants, to any of the
+parties who would receive the covered work from you, a discriminatory
+patent license (a) in connection with copies of the covered work
+conveyed by you (or copies made from those copies), or (b) primarily
+for and in connection with specific products or compilations that
+contain the covered work, unless you entered into that arrangement,
+or that patent license was granted, prior to 28 March 2007.
+
+  Nothing in this License shall be construed as excluding or limiting
+any implied license or other defenses to infringement that may
+otherwise be available to you under applicable patent law.
+
+  12. No Surrender of Others' Freedom.
+
+  If conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License.  If you cannot convey a
+covered work so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you may
+not convey it at all.  For example, if you agree to terms that obligate you
+to collect a royalty for further conveying from those to whom you convey
+the Program, the only way you could satisfy both those terms and this
+License would be to refrain entirely from conveying the Program.
+
+  13. Remote Network Interaction; Use with the GNU General Public License.
+
+  Notwithstanding any other provision of this License, if you modify the
+Program, your modified version must prominently offer all users
+interacting with it remotely through a computer network (if your version
+supports such interaction) an opportunity to receive the Corresponding
+Source of your version by providing access to the Corresponding Source
+from a network server at no charge, through some standard or customary
+means of facilitating copying of software.  This Corresponding Source
+shall include the Corresponding Source for any work covered by version 3
+of the GNU General Public License that is incorporated pursuant to the
+following paragraph.
+
+  Notwithstanding any other provision of this License, you have
+permission to link or combine any covered work with a work licensed
+under version 3 of the GNU General Public License into a single
+combined work, and to convey the resulting work.  The terms of this
+License will continue to apply to the part which is the covered work,
+but the work with which it is combined will remain governed by version
+3 of the GNU General Public License.
+
+  14. Revised Versions of this License.
+
+  The Free Software Foundation may publish revised and/or new versions of
+the GNU Affero General Public License from time to time.  Such new versions
+will be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+  Each version is given a distinguishing version number.  If the
+Program specifies that a certain numbered version of the GNU Affero General
+Public License "or any later version" applies to it, you have the
+option of following the terms and conditions either of that numbered
+version or of any later version published by the Free Software
+Foundation.  If the Program does not specify a version number of the
+GNU Affero General Public License, you may choose any version ever published
+by the Free Software Foundation.
+
+  If the Program specifies that a proxy can decide which future
+versions of the GNU Affero General Public License can be used, that proxy's
+public statement of acceptance of a version permanently authorizes you
+to choose that version for the Program.
+
+  Later license versions may give you additional or different
+permissions.  However, no additional obligations are imposed on any
+author or copyright holder as a result of your choosing to follow a
+later version.
+
+  15. Disclaimer of Warranty.
+
+  THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
+APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
+HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
+OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
+THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
+IS WITH YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
+ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+  16. Limitation of Liability.
+
+  IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
+THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
+GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
+USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
+DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
+PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
+EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGES.
+
+  17. Interpretation of Sections 15 and 16.
+
+  If the disclaimer of warranty and limitation of liability provided
+above cannot be given local legal effect according to their terms,
+reviewing courts shall apply local law that most closely approximates
+an absolute waiver of all civil liability in connection with the
+Program, unless a warranty or assumption of liability accompanies a
+copy of the Program in return for a fee.
+
+                     END OF TERMS AND CONDITIONS
+
+            How to Apply These Terms to Your New Programs
+
+  If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+  To do so, attach the following notices to the program.  It is safest
+to attach them to the start of each source file to most effectively
+state the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+    <one line to give the program's name and a brief idea of what it does.>
+    Copyright (C) <year>  <name of author>
+
+    This program is free software: you can redistribute it and/or modify
+    it under the terms of the GNU Affero General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU Affero General Public License for more details.
+
+    You should have received a copy of the GNU Affero General Public License
+    along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+  If your software can interact with users remotely through a computer
+network, you should also make sure that it provides a way for users to
+get its source.  For example, if your program is a web application, its
+interface could display a "Source" link that leads users to an archive
+of the code.  There are many ways you could offer source, and different
+solutions will be better for different programs; see section 13 for the
+specific requirements.
+
+  You should also get your employer (if you work as a programmer) or school,
+if any, to sign a "copyright disclaimer" for the program, if necessary.
+For more information on this, and how to apply and follow the GNU AGPL, see
+<https://www.gnu.org/licenses/>.

build.gradle

@@ -4,7 +4,7 @@ plugins {
 }
 
 group = 'com'
-version = '1.0'
+version = '2.8'
 
 repositories {
     mavenCentral()
@@ -16,18 +16,33 @@ repositories {
         name = "sonatype"
         url = "https://oss.sonatype.org/content/groups/public/"
     }
+    maven {
+        url = 'https://repo.codemc.io/repository/maven-public/'
+    }
 }
 
 dependencies {
     compileOnly "org.spigotmc:spigot-api:1.13-R0.1-SNAPSHOT"
-    implementation 'com.squareup.okhttp3:okhttp:4.9.3'
+    implementation 'org.bstats:bstats-bukkit:3.0.2'
+    implementation('org.jodd:jodd-http:6.3.0') {
+        exclude group: 'org.jodd', module: 'jodd-core'
+        exclude group: 'org.jodd', module: 'jodd-bean'
+    }
     implementation 'org.json:json:20231013'
 }
 
 shadowJar {
     archiveFileName = "MineChatGPT-${project.version}.jar"
-    relocate 'okhttp3', 'com.ddaodan.minechatgpt.libs.okhttp3'
+    minimize()
+    mergeServiceFiles()
+
+    exclude 'META-INF/**'
+    exclude '**/*.properties'
+    exclude '**/jodd/http/upload/**'
+    
+    relocate 'jodd', 'com.ddaodan.shaded.jodd'
     relocate 'org.json', 'com.ddaodan.minechatgpt.libs.org.json'
+    relocate 'org.bstats', 'com.ddaodan.minechatgpt.libs.org.bstats'
 }
 
 def targetJavaVersion = 8

changelog.md

@@ -1,3 +1,33 @@
 # 更新日志
+## 2.7
+- 多语言支持
+- 配置文件添加双语注释
+- 新增负载均衡，可轮询或随机选择API key
+- 优化构建配置，插件体积减小
+- 项目规范化
+在此感谢Trae，以上功能都是靠它才能实现的！
+## 2.6
+- 添加角色功能
+- 移除prompt功能
+## 2.5
+- 允许插件在Folia加载
+- 修复 #4：[BUG]部分情况下回复乱码
+## 2.4
+- 添加自定义prompt
+- 移除了没有任何效果的配置文件自动更新功能
+## 2.3
+- 修复 #3：[BUG]提问乱码/答非所问
+## 2.2
+- 修复插件无法加载
+- 修复 #2：模型切换实际并不生效
+## 2.1
+- 接入bStats
+- 调整配置文件，添加更新功能，现在更新插件应该不用再备份配置文件了
+- 修改了gradle中错误的依赖
+## 2.0
+- 更换为jodd-http库，减少插件大小
+- 支持指令补全
+- 现在可以在配置文件中修改所有消息的颜色了
+- 默认配置文件中增加了一些模型
 ## 1.0
 - 插件发布

gradle.properties

@@ -0,0 +1,2 @@
+systemProp.org.gradle.internal.http.connectionTimeout=60000
+systemProp.org.gradle.internal.http.socketTimeout=120000

gradle/wrapper/gradle-wrapper.properties

@@ -1,7 +1,7 @@
 distributionBase=GRADLE_USER_HOME
 distributionPath=wrapper/dists
 distributionUrl=https\://services.gradle.org/distributions/gradle-8.7-bin.zip
-networkTimeout=10000
+networkTimeout=60000
 validateDistributionUrl=true
 zipStoreBase=GRADLE_USER_HOME
 zipStorePath=wrapper/dists

readme.md

@@ -1,60 +1,108 @@
 # MineChatGPT
+![minechatgpt](https://socialify.git.ci/ddaodan/minechatgpt/image?description=1&descriptionEditable=%E5%9C%A8Minecraft%E4%B8%AD%E4%B8%8EChatGPT%E4%BA%A4%E6%B5%81&font=Inter&issues=1&language=1&name=1&pattern=Solid&stargazers=1&theme=Auto)  
+[![modrinth-gallery](https://cdn.jsdelivr.net/npm/@intergrav/devins-badges@3/assets/cozy-minimal/documentation/modrinth-gallery_vector.svg)](https://modrinth.com/plugin/minechatgpt) [![spigot](https://cdn.jsdelivr.net/npm/@intergrav/devins-badges@3/assets/cozy-minimal/supported/spigot_vector.svg)](https://www.spigotmc.org/resources/minechatgpt.118963/) ![paper](https://cdn.jsdelivr.net/npm/@intergrav/devins-badges@3/assets/cozy-minimal/supported/paper_vector.svg) ![java8](https://cdn.jsdelivr.net/npm/@intergrav/devins-badges@3/assets/cozy-minimal/built-with/java8_vector.svg)  
+![GitHub Release](https://img.shields.io/github/v/release/ddaodan/minechatgpt?label=version) ![bStats Servers](https://img.shields.io/bstats/servers/22635) ![bStats Players](https://img.shields.io/bstats/players/22635) [![Modrinth Downloads](https://img.shields.io/modrinth/dt/Op2X2eDG?logo=modrinth)](https://modrinth.com/plugin/minechatgpt/versions) [![Spiget Downloads](https://img.shields.io/spiget/downloads/118963?logo=spigotmc)](https://www.spigotmc.org/resources/minechatgpt.118963/)
+
 在Minecraft中与ChatGPT交流  
 
 所有的代码都是ChatGPT写的哦
 
 ## 功能
-- [x] OpenAPI格式
+- OpenAPI格式
-- [x] 自定义模型
+- 自定义模型
-- [x] ChatGPT反代
+- ChatGPT反代
-- [ ] 指令补全
+- 负载均衡
-- [ ] 上下文对话
+- 指令补全
-- [ ] 自定义prompt
+- 上下文对话
+- 多角色
+- Folia支持
 
 ## 安装
-- 下载插件，放在plugins文件夹中
+1. 下载插件，放在plugins文件夹中
-- 重启服务器
+2. 重启服务器
-
+  > 为兼容更多版本，插件没有规定Bukkit API version，因此在较高版本加载插件时，控制台可能会出现以下错误信息，这属于正常现象。  
-## 配置文件`config.yml`
+  > ```
+  > [Server thread/WARN]: Initializing Legacy Material Support. Unless you have legacy plugins and/or data this is a bug!
+  > [Server thread/WARN]: Legacy plugin MineChatGPT v1.0 does not specify an api-version.
+  > ```
+3. 打开配置文件`config.yml`，修改以下设置：
   ```yaml
-# API 相关设置
+  # ======================================================
+  # API Configuration
+  # API 设置
+  # ======================================================
   api:
+    # Your OpenAI API keys, used for authentication
+    # To obtain an API key, visit https://platform.openai.com/account/api-keys and create a new API key
     # 你的 OpenAI API key，用于身份验证
-  # 获取 API key 的方法：访问 //platform.openai.com/account/api-keys 并创建一个新的 API key
+    # 获取 API key 的方法：访问 https://platform.openai.com/account/api-keys 并创建一个新的 API key
-  key: "your_openai_api_key"
+    keys:
-  # OpenAI API 的基础 URL，用于构建请求
+      - "sk-your_openai_api_key_1"
-  base_url: "https://api.openai.com/v1"
+      # You can add multiple API keys below
-# 支持的模型列表
+      # 可以添加多个API key
-models:
+      # - "sk-your_openai_api_key_2"
-  # OpenAI ChatGPT
+      # - "sk-your_openai_api_key_3"
-  - "gpt-3.5-turbo"
-  - "gpt-4"
-# 默认使用的模型
-default_model: "gpt-3.5-turbo"
-# 消息相关设置
-messages:
-  reload: "已重新加载配置文件！"
-  help: "===== MineChatGPT 帮助 ====="
-  help_ask: "/chatgpt <text> - 向ChatGPT提问"
-  help_reload: "/chatgpt reload - 重新加载配置文件"
-  help_model: "/chatgpt model <model_name> - 切换至其他模型"
-  help_modellist: "/chatgpt modellist - 可用的模型列表"
-  usage: "输入： /chatgpt model <model_name>"
-  model_switch: "已切换至模型 %s"
-  chatgpt_error: "无法联系ChatGPT。"
-  chatgpt_response: "ChatGPT: %s"
-  question: "你: %s"
-  invalid_model: "模型无效。使用 /chatgpt modellist 查看可用模型。"
-  available_models: "可用模型列表："
-  no_permission: "你没有权限使用这个指令。需要的权限：%s"
-```
     
+    # API key selection method: "round_robin" or "random"
+    # Round Robin: Use each API key in turn
+    # Random: Randomly select an API key
+    # API key 选择方法："round_robin"（轮询）或 "random"（随机）
+    # 轮询：依次使用每个API key
+    # 随机：随机选择一个API key
+    selection_method: "round_robin"
+    
+    # The base URL for the OpenAI API, used to construct requests
+    # If you cannot access the official API, you can use a proxy service
+    # OpenAI API 的基础 URL，用于构建请求
+    # 如果你无法访问官方API，可以使用代理服务
+    base_url: "https://api.openai.com/v1"
+  ```
+4. 在控制台中输入`/chatgpt reload`重新加载配置文件
+
+## 截图
+- 服务端截图（Spigot 1.20.1）
+![](https://i.ddaodan.cc/images/CWindowssystem32cmd.exe_20240712406.png)
+- 插件截图
+![](https://i.ddaodan.cc/images/Minecraft_1.20.1_-__20240712407.png)
+- 对话截图（使用FastGPT训练的自定义知识库）
+![](https://i.ddaodan.cc/images/Minecraft_1.20.1_-__20240712408.png)
+
+## 指令与权限
+|指令|权限|描述|
+|-|-|-|
+|`/chatgpt`|minechatgpt.use|查看插件帮助|
+|`/chatgpt <text>`|minechatgpt.use|向ChatGPT提问|
+|`/chatgpt reload`|minechatgpt.reload|重新加载配置文件|
+|`/chatgpt model <model_name>`|minechatgpt.model|切换至其他模型|
+|`/chatgpt modellist`|minechatgpt.modellist|查看可用的模型列表|
+|`/chatgpt context`|minechatgpt.context|切换连续对话模式|
+|`/chatgpt clear`|minechatgpt.clear|清空对话历史|
+|`/chatgpt character [character_name]`|minechatgpt.character|列出或切换角色|
+
+## 兼容的版本
+只列出经过测试的版本
+
+|服务端|支持情况|
+|-|-|
+|Luminol 1.21|✔ 支持|
+|Mohist 1.20.1|✔ 支持|
+|Spigot 1.20.1|✔ 支持|
+|Spigot 1.12.2|✔ 支持|
+|KCauldron 1.7.10|× 不支持|
 
 ## 常见问题
-### `Failed to contact ChatGPT.` `无法联系ChatGPT。`
+### 提问后显示`Failed to contact ChatGPT.` `无法联系ChatGPT。`
 检查控制台输出的错误内容。
-### `connect timeout` `connect reset`
+### 提问后控制台有`connect timeout` `connect reset`等类似的提示
 检查`config.yml`中的`base_url`能否正常访问。如果你无法连接到OpenAI官方的API地址，可以考虑使用其他反代。
-
+### 我可以添加其他模型吗？
+可以，只要模型支持OpenAI的API，就可以使用。  
+不推荐使用推理模型，因为推理模型的响应时间较长，在没有流式响应的情况下，玩家会认为插件出现了问题。
+### 我没有ChatGPT的账号，可以用吗？
+可以，目前有很多代理网站，可以很轻松地使用，而且还支持其他模型，费用通常来说也会比官方便宜。如果你愿意，也可以使用我的代理，目前仅在我的QQ群：226385797中提供。
+### 是否会支持Folia
+理论上插件可以在Folia上运行，但插件的代码并没有针对Folia进行过优化，因此可能会有一些问题。如果你愿意，可以尝试使用Folia运行插件，但不保证插件可以正常运行。
 ## 赞助
-![afdian-ddaodan.jpeg](https://i.ddaodan.cn/images/afdian-ddaodan.jpeg)
+[![](https://i.ddaodan.cc/images/afdian-ddaodan.jpeg)](https://afdian.com/a/ddaodan)
+## 统计
+[![](https://bstats.org/signatures/bukkit/MineChatGPT.svg)](https://bstats.org/plugin/bukkit/MineChatGPT/22635)

src/main/java/com/ddaodan/MineChatGPT/CommandHandler.java

@@ -1,140 +1,52 @@
 package com.ddaodan.MineChatGPT;
 
+import com.ddaodan.MineChatGPT.service.ApiService;
+import com.ddaodan.MineChatGPT.service.CommandService;
+import com.ddaodan.MineChatGPT.service.RequestCoordinator;
+import com.ddaodan.MineChatGPT.service.UserSessionManager;
 import org.bukkit.command.Command;
 import org.bukkit.command.CommandExecutor;
 import org.bukkit.command.CommandSender;
-import org.bukkit.ChatColor;
-import okhttp3.*;
-import org.json.JSONArray;
-import org.json.JSONObject;
-
-import java.io.IOException;
-import java.util.logging.Level;
-import java.util.logging.Logger;
-import java.util.List;
-
 public class CommandHandler implements CommandExecutor {
-    private final Main plugin;
     private final ConfigManager configManager;
-    private static final Logger logger = Logger.getLogger(CommandHandler.class.getName());
+    private final CommandService commandService;
+    private final UserSessionManager sessionManager;
 
-    public CommandHandler(Main plugin, ConfigManager configManager) {
+    public CommandHandler(ConfigManager configManager, UserSessionManager sessionManager, RequestCoordinator requestCoordinator) {
-        this.plugin = plugin;
         this.configManager = configManager;
+        this.sessionManager = sessionManager;
+        this.commandService = new CommandService(configManager, requestCoordinator, sessionManager);
     }
 
     @Override
     public boolean onCommand(CommandSender sender, Command command, String label, String[] args) {
+        String userId = sender.getName();
+
         if (command.getName().equalsIgnoreCase("chatgpt")) {
             if (args.length == 0) {
-                sendHelpMessage(sender);
+                commandService.sendHelpMessage(sender);
                 return true;
             }
             
             String subCommand = args[0];
             if (subCommand.equalsIgnoreCase("reload")) {
-                if (!sender.hasPermission("minechatgpt.reload")) {
+                return commandService.handleReloadCommand(sender);
-                    sender.sendMessage(ChatColor.RED + String.format(configManager.getNoPermissionMessage(), "minechatgpt.reload"));
-                    return true;
-                }
-                configManager.reloadConfig();
-                sender.sendMessage(ChatColor.GREEN + configManager.getReloadMessage());
-                return true;
             } else if (subCommand.equalsIgnoreCase("model")) {
-                if (!sender.hasPermission("minechatgpt.model")) {
+                return commandService.handleModelCommand(sender, args);
-                    sender.sendMessage(ChatColor.RED + String.format(configManager.getNoPermissionMessage(), "minechatgpt.model"));
-                    return true;
-                }
-                if (args.length < 2) {
-                    sender.sendMessage(ChatColor.RED + configManager.getUsageMessage());
-                    return true;
-                }
-                String model = args[1];
-                List<String> models = configManager.getModels();
-                if (models.contains(model)) {
-                    // Logic to switch model
-                    sender.sendMessage(ChatColor.GREEN + String.format(configManager.getModelSwitchMessage(), model));
-                } else {
-                    sender.sendMessage(ChatColor.RED + configManager.getInvalidModelMessage());
-                }
-                return true;
             } else if (subCommand.equalsIgnoreCase("modellist")) {
-                if (!sender.hasPermission("minechatgpt.modellist")) {
+                return commandService.handleModelListCommand(sender);
-                    sender.sendMessage(ChatColor.RED + String.format(configManager.getNoPermissionMessage(), "minechatgpt.modellist"));
+            } else if (subCommand.equalsIgnoreCase("context")) {
-                    return true;
+                return commandService.handleContextCommand(sender, userId);
-                }
+            } else if (subCommand.equalsIgnoreCase("clear")) {
-                List<String> models = configManager.getModels();
+                return commandService.handleClearCommand(sender, userId);
-                sender.sendMessage(ChatColor.YELLOW + configManager.getAvailableModelsMessage());
+            } else if (subCommand.equalsIgnoreCase("character")) {
-                for (String model : models) {
+                return commandService.handleCharacterCommand(sender, args, userId);
-                    sender.sendMessage(ChatColor.YELLOW + "- " + model);
+            } else if (subCommand.equalsIgnoreCase("stats")) {
-                }
+                return commandService.handleStatsCommand(sender, args, userId);
-                return true;
             } else {
-                if (!sender.hasPermission("minechatgpt.use")) {
+                return commandService.handleAskCommand(sender, args, userId);
-                    sender.sendMessage(ChatColor.RED + String.format(configManager.getNoPermissionMessage(), "minechatgpt.use"));
-                    return true;
-                }
-                String question = String.join(" ", args);
-                // Logic to send question to ChatGPT
-                sender.sendMessage(ChatColor.AQUA + String.format(configManager.getQuestionMessage(), question));
-                askChatGPT(sender, question);
-                return true;
             }
         }
         return false;
     }
-
-    private void askChatGPT(CommandSender sender, String question) {
-        OkHttpClient client = new OkHttpClient();
-
-        MediaType mediaType = MediaType.parse("application/json");
-        JSONObject json = new JSONObject();
-        json.put("model", configManager.getDefaultModel());
-        JSONArray messages = new JSONArray();
-        JSONObject message = new JSONObject();
-        message.put("role", "user");
-        message.put("content", question);
-        messages.put(message);
-        json.put("messages", messages);
-
-        RequestBody body = RequestBody.create(mediaType, json.toString());
-        Request request = new Request.Builder()
-                .url(configManager.getBaseUrl() + "/chat/completions")
-                .post(body)
-                .addHeader("Content-Type", "application/json")
-                .addHeader("Authorization", "Bearer " + configManager.getApiKey())
-                .build();
-
-        client.newCall(request).enqueue(new Callback() {
-            @Override
-            public void onFailure(Call call, IOException e) {
-                e.printStackTrace();
-                logger.log(Level.SEVERE, "Failed to contact ChatGPT: " + e.getMessage(), e);
-                sender.sendMessage(ChatColor.RED + configManager.getChatGPTErrorMessage());
-            }
-
-            @Override
-            public void onResponse(Call call, Response response) throws IOException {
-                if (response.isSuccessful()) {
-                    String responseBody = response.body().string();
-                    JSONObject jsonResponse = new JSONObject(responseBody);
-                    String answer = jsonResponse.getJSONArray("choices").getJSONObject(0).getJSONObject("message").getString("content");
-                    sender.sendMessage(ChatColor.AQUA + String.format(configManager.getChatGPTResponseMessage(), answer));
-                } else {
-                    String errorBody = response.body().string();
-                    logger.log(Level.SEVERE, "Failed to get a response from ChatGPT: " + errorBody);
-                    sender.sendMessage(ChatColor.RED + configManager.getChatGPTErrorMessage());
-                }
-                response.close();
-            }
-        });
-    }
-
-    private void sendHelpMessage(CommandSender sender) {
-        sender.sendMessage(ChatColor.YELLOW + configManager.getHelpMessage());
-        sender.sendMessage(ChatColor.YELLOW + configManager.getHelpAskMessage());
-        sender.sendMessage(ChatColor.YELLOW + configManager.getHelpReloadMessage());
-        sender.sendMessage(ChatColor.YELLOW + configManager.getHelpModelMessage());
-        sender.sendMessage(ChatColor.YELLOW + configManager.getHelpModelListMessage());
-    }
 }

src/main/java/com/ddaodan/MineChatGPT/ConfigManager.java

@@ -1,91 +1,334 @@
 package com.ddaodan.MineChatGPT;
 
+import com.ddaodan.MineChatGPT.util.ConfigFileUpdater;
+import org.bukkit.ChatColor;
+import org.bukkit.configuration.ConfigurationSection;
 import org.bukkit.configuration.file.FileConfiguration;
+
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.LinkedHashMap;
 import java.util.List;
+import java.util.Map;
+import java.util.concurrent.atomic.AtomicInteger;
 
 public class ConfigManager {
     private final Main plugin;
     private FileConfiguration config;
+    private String currentModel;
+    private LanguageManager languageManager;
+    private volatile Map<String, String> charactersCache;
 
     public ConfigManager(Main plugin) {
         this.plugin = plugin;
         reloadConfig();
+        // 获取语言设置
+        String language = config.getString("language", "en");
+        this.languageManager = new LanguageManager(plugin, language);
+        this.charactersCache = null;
+    }
+    public boolean isDebugMode() {
+        return config.getBoolean("debug", false);
     }
-
     public void reloadConfig() {
+        ConfigFileUpdater.UpdateResult updateResult = ConfigFileUpdater.updateIfMissingKeys(plugin, "config.yml");
+        if (updateResult.updated) {
+            plugin.getLogger().info("Config updated: inserted " + updateResult.insertedPaths
+                    + " missing path(s). Backup: " + updateResult.backupFileName);
+        }
         plugin.reloadConfig();
         config = plugin.getConfig();
+        currentModel = config.getString("default_model");
+        charactersCache = null;
+        
+        // 重新加载语言文件
+        if (languageManager != null) {
+            String language = config.getString("language", "en");
+            languageManager.setLanguage(language);
+        }
     }
 
+    private String translateColorCodes(String message) {
+        return ChatColor.translateAlternateColorCodes('&', message);
+    }
+    public String getCurrentModel() {
+        return currentModel;
+    }
+    public void setCurrentModel(String model) {
+        currentModel = model;
+    }
+    private final AtomicInteger currentKeyIndex = new AtomicInteger(0);
+    
     public String getApiKey() {
-        return config.getString("api.key");
+        List<String> keys = new ArrayList<>();
+        for (String key : config.getStringList("api.keys")) {
+            if (isUsableApiKey(key)) {
+                keys.add(key);
+            }
+        }
+        if (keys.isEmpty()) {
+            String legacyKey = config.getString("api.key");
+            if (isUsableApiKey(legacyKey)) {
+                return legacyKey;
+            }
+            return "";
+        }
+
+        String selectionMethod = config.getString("api.selection_method", "round_robin");
+
+        if ("random".equalsIgnoreCase(selectionMethod)) {
+            int randomIndex = (int) (Math.random() * keys.size());
+            return keys.get(randomIndex);
+        } else {
+            int index = Math.floorMod(currentKeyIndex.getAndIncrement(), keys.size());
+            return keys.get(index);
+        }
+    }
+
+    private boolean isUsableApiKey(String key) {
+        if (key == null) {
+            return false;
+        }
+        String trimmed = key.trim();
+        if (trimmed.isEmpty()) {
+            return false;
+        }
+        return !trimmed.startsWith("sk-your_openai_api_key_");
     }
 
     public String getBaseUrl() {
-        return config.getString("api.base_url");
+        return config.getString("api.base_url", "https://api.openai.com/v1");
+    }
+    public int getApiConnectTimeoutMs() {
+        return config.getInt("api.connect_timeout_ms", 10000);
+    }
+    public int getApiTimeoutMs() {
+        return config.getInt("api.timeout_ms", 30000);
+    }
+    public int getApiThreadPoolSize() {
+        return Math.max(1, config.getInt("api.thread_pool_size", 4));
     }
-
     public String getDefaultModel() {
         return config.getString("default_model");
     }
-
     public String getReloadMessage() {
-        return config.getString("messages.reload");
+        return languageManager.getMessage("reload");
     }
-
     public List<String> getModels() {
         return config.getStringList("models");
     }
-
     public String getHelpMessage() {
-        return config.getString("messages.help");
+        return languageManager.getMessage("help");
     }
-
     public String getHelpAskMessage() {
-        return config.getString("messages.help_ask");
+        return languageManager.getMessage("help_ask");
     }
-
-    public String getHelpModelListMessage() {
-        return config.getString("messages.help_modellist");
-    }
-
     public String getHelpReloadMessage() {
-        return config.getString("messages.help_reload");
+        return languageManager.getMessage("help_reload");
     }
-
     public String getHelpModelMessage() {
-        return config.getString("messages.help_model");
+        return languageManager.getMessage("help_model");
     }
-
+    public String getHelpModelListMessage() {
-    public String getUsageMessage() {
+        return languageManager.getMessage("help_modellist");
-        return config.getString("messages.usage");
+    }
+    public String getHelpContextMessage() {
+        return languageManager.getMessage("help_context");
+    }
+    public String getHelpClearMessage() {
+        return languageManager.getMessage("help_clear");
+    }
+    public String getHelpCharacterMessage() {
+        return languageManager.getMessage("help_character");
+    }
+    public String getHelpStatsMessage() {
+        return languageManager.getMessage("help_stats");
     }
-
     public String getModelSwitchMessage() {
-        return config.getString("messages.model_switch");
+        return languageManager.getMessage("model_switch");
     }
-
     public String getChatGPTErrorMessage() {
-        return config.getString("messages.chatgpt_error");
+        return languageManager.getMessage("chatgpt_error");
+    }
+    public String getNoApiKeyMessage() {
+        return languageManager.getMessage("no_api_key", "&cNo API key configured. Please set api.keys in config.yml.");
     }
-
     public String getChatGPTResponseMessage() {
-        return config.getString("messages.chatgpt_response");
+        return languageManager.getMessage("chatgpt_response", "&b%s: %s");
     }
-
     public String getQuestionMessage() {
-        return config.getString("messages.question");
+        return languageManager.getMessage("question");
+    }
+    public String getQueuedMessage() {
+        return languageManager.getMessage("queued", "&eYour request has been queued. Position: %s");
+    }
+    public String getQueueFullMessage() {
+        return languageManager.getMessage("queue_full", "&cQueue is full. Please try again later.");
+    }
+    public String getQueueFullUserMessage() {
+        return languageManager.getMessage("queue_full_user", "&cYou have too many pending requests. Please wait.");
+    }
+    public String getCooldownMessage() {
+        return languageManager.getMessage("cooldown", "&cYou're sending requests too fast. Please wait %s ms.");
+    }
+    public String getRateLimitedMessage() {
+        return languageManager.getMessage("rate_limited", "&cRate limited. Please try again later.");
+    }
+    public String getStatsHeaderMessage() {
+        return languageManager.getMessage("stats_header", "&e===== MineChatGPT Stats =====");
+    }
+    public String getStatsGlobalMessage() {
+        return languageManager.getMessage("stats_global", "&eGlobal tokens: %s (prompt=%s, completion=%s), requests=%s");
+    }
+    public String getStatsUserMessage() {
+        return languageManager.getMessage("stats_user", "&eYour tokens: %s (prompt=%s, completion=%s), requests=%s");
+    }
+    public String getStatsResetMessage() {
+        return languageManager.getMessage("stats_reset", "&aStats reset.");
     }
-
     public String getInvalidModelMessage() {
-        return config.getString("messages.invalid_model");
+        return languageManager.getMessage("invalid_model");
     }
-
     public String getAvailableModelsMessage() {
-        return config.getString("messages.available_models");
+        return languageManager.getMessage("available_models");
+    }
+    public String getNoPermissionMessage() {
+        return languageManager.getMessage("no_permission");
+    }
+    public String getCurrentModelInfoMessage() {
+        return languageManager.getMessage("current_model_info");
+    }
+    public int getMaxHistorySize() {
+        return config.getInt("conversation.max_history_size", 10);
+    }
+    public int getMaxContextTokens() {
+        return config.getInt("conversation.max_context_tokens", 2000);
+    }
+    public int getReserveCompletionTokens() {
+        return config.getInt("conversation.reserve_completion_tokens", 400);
+    }
+    public boolean isSummarizationEnabled() {
+        return config.getBoolean("conversation.summarization.enabled", true);
+    }
+    public String getSummarizationModel() {
+        return config.getString("conversation.summarization.model", getCurrentModel());
+    }
+    public int getSummarizationTriggerTokens() {
+        return config.getInt("conversation.summarization.trigger_tokens", 1800);
+    }
+    public int getSummarizationKeepLastMessages() {
+        return Math.max(0, config.getInt("conversation.summarization.keep_last_messages", 6));
+    }
+    public boolean isContextEnabled() {
+        return config.getBoolean("conversation.context_enabled", false);
+    }
+    public String getContextToggleMessage() {
+        return languageManager.getMessage("context_toggle", "&eContext is now %s.");
+    }
+    public String getContextToggleEnabledMessage() {
+        return languageManager.getMessage("context_toggle_enabled", "&aenabled");
+    }
+    public String getContextToggleDisabledMessage() {
+        return languageManager.getMessage("context_toggle_disabled", "&edisabled");
+    }
+    public String getClearMessage() {
+        return languageManager.getMessage("clear", "&aConversation history has been cleared.");
+    }
+    public String getCharacterSwitchedMessage() {
+        return languageManager.getMessage("character_switched", "&aSwitched to character: %s");
+    }
+    public String getAvailableCharactersMessage() {
+        return languageManager.getMessage("available_characters", "&eAvailable characters:");
+    }
+    public String getInvalidCharacterMessage() {
+        return languageManager.getMessage("invalid_character", "&cInvalid character. Use /chatgpt character to list available characters.");
+    }
+    public Map<String, String> getCharacters() {
+        Map<String, String> cached = charactersCache;
+        if (cached != null) {
+            return cached;
         }
 
-    public String getNoPermissionMessage() {
+        ConfigurationSection section = config.getConfigurationSection("characters");
-        return config.getString("messages.no_permission");
+        if (section == null) {
+            charactersCache = Collections.emptyMap();
+            return charactersCache;
+        }
+
+        Map<String, String> characters = new LinkedHashMap<>();
+        for (String key : section.getKeys(false)) {
+            String prompt = section.getString(key);
+            if (prompt != null) {
+                characters.put(key, prompt);
+            }
+        }
+
+        charactersCache = Collections.unmodifiableMap(characters);
+        return charactersCache;
+    }
+    public String getCurrentCharacter(String userId) {
+        return config.getString("users." + userId + ".character", "ChatGPT");
+    }
+    public void setCurrentCharacter(String userId, String character) {
+        config.set("users." + userId + ".character", character);
+        plugin.saveConfig();
+    }
+
+    public boolean isRateLimitEnabled() {
+        return config.getBoolean("rate_limit.enabled", true);
+    }
+    public String getRateLimitMode() {
+        return config.getString("rate_limit.mode", "both");
+    }
+    public long getRateLimitCooldownMs() {
+        return Math.max(0L, config.getLong("rate_limit.cooldown_ms", 1000L));
+    }
+    public String getRateLimitTokenEstimator() {
+        return config.getString("rate_limit.token_estimator", "approx_chars");
+    }
+    public int getRateLimitAssumedCompletionTokens() {
+        return Math.max(0, config.getInt("rate_limit.assumed_completion_tokens", 300));
+    }
+
+    public int getPerUserRequestsPerMinute() {
+        return Math.max(0, config.getInt("rate_limit.per_user.requests_per_minute", 6));
+    }
+    public int getPerUserBurstRequests() {
+        return Math.max(0, config.getInt("rate_limit.per_user.burst_requests", 3));
+    }
+    public int getPerUserTokensPerMinute() {
+        return Math.max(0, config.getInt("rate_limit.per_user.tokens_per_minute", 4000));
+    }
+    public int getPerUserBurstTokens() {
+        return Math.max(0, config.getInt("rate_limit.per_user.burst_tokens", 2000));
+    }
+
+    public int getGlobalRequestsPerMinute() {
+        return Math.max(0, config.getInt("rate_limit.global.requests_per_minute", 60));
+    }
+    public int getGlobalBurstRequests() {
+        return Math.max(0, config.getInt("rate_limit.global.burst_requests", 20));
+    }
+    public int getGlobalTokensPerMinute() {
+        return Math.max(0, config.getInt("rate_limit.global.tokens_per_minute", 100000));
+    }
+    public int getGlobalBurstTokens() {
+        return Math.max(0, config.getInt("rate_limit.global.burst_tokens", 20000));
+    }
+
+    public boolean isQueueEnabled() {
+        return config.getBoolean("queue.enabled", true);
+    }
+    public int getQueueMaxSize() {
+        return Math.max(0, config.getInt("queue.max_size", 100));
+    }
+    public int getQueueMaxPerUser() {
+        return Math.max(0, config.getInt("queue.max_per_user", 3));
+    }
+    public int getQueueMaxInFlight() {
+        return Math.max(1, config.getInt("queue.max_in_flight", getApiThreadPoolSize()));
+    }
+    public int getQueueDispatchPerTick() {
+        return Math.max(1, config.getInt("queue.dispatch_per_tick", 2));
     }
 }

src/main/java/com/ddaodan/MineChatGPT/ConversationContext.java

@@ -0,0 +1,84 @@
+package com.ddaodan.MineChatGPT;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Deque;
+import java.util.List;
+
+public class ConversationContext {
+    public static final class Message {
+        private final String role;
+        private final String content;
+
+        public Message(String role, String content) {
+            this.role = role;
+            this.content = content;
+        }
+
+        public String getRole() {
+            return role;
+        }
+
+        public String getContent() {
+            return content;
+        }
+    }
+
+    private final Deque<Message> history;
+    private final int maxHistorySize;
+
+    public ConversationContext(int maxHistorySize) {
+        this.maxHistorySize = maxHistorySize;
+        this.history = new ArrayDeque<>();
+    }
+
+    public void addUserMessage(String message) {
+        addMessage("user", message);
+    }
+
+    public void addAssistantMessage(String message) {
+        addMessage("assistant", message);
+    }
+
+    private void addMessage(String role, String message) {
+        if (history.size() >= maxHistorySize) {
+            history.pollFirst();
+        }
+        history.addLast(new Message(role, message));
+    }
+
+    public List<Message> getMessages() {
+        return new ArrayList<>(history);
+    }
+
+    public void setMessages(List<Message> messages) {
+        history.clear();
+        if (messages == null || messages.isEmpty()) {
+            return;
+        }
+        int startIndex = Math.max(0, messages.size() - maxHistorySize);
+        for (int i = startIndex; i < messages.size(); i++) {
+            Message msg = messages.get(i);
+            if (msg != null) {
+                history.addLast(new Message(msg.getRole(), msg.getContent()));
+            }
+        }
+    }
+
+    public void trimToLast(int keepLast) {
+        if (keepLast < 0) {
+            keepLast = 0;
+        }
+        while (history.size() > keepLast) {
+            history.pollFirst();
+        }
+    }
+
+    public void removeOldestMessage() {
+        history.pollFirst();
+    }
+
+    public void clearHistory() {
+        history.clear();
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/LanguageManager.java

@@ -0,0 +1,73 @@
+package com.ddaodan.MineChatGPT;
+
+import com.ddaodan.MineChatGPT.util.ConfigFileUpdater;
+import org.bukkit.ChatColor;
+import org.bukkit.configuration.file.FileConfiguration;
+import org.bukkit.configuration.file.YamlConfiguration;
+
+import java.io.File;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.nio.charset.StandardCharsets;
+
+public class LanguageManager {
+    private final Main plugin;
+    private FileConfiguration langConfig;
+    private String currentLanguage;
+    private File langFile;
+
+    public LanguageManager(Main plugin, String language) {
+        this.plugin = plugin;
+        this.currentLanguage = language;
+        loadLanguage();
+    }
+
+    public void loadLanguage() {
+        String resourcePath = "lang/" + currentLanguage + ".yml";
+        langFile = new File(plugin.getDataFolder(), "lang" + File.separator + currentLanguage + ".yml");
+
+        if (plugin.getResource(resourcePath) != null) {
+            ConfigFileUpdater.UpdateResult updateResult = ConfigFileUpdater.updateIfMissingKeys(plugin, resourcePath);
+            if (updateResult.updated) {
+                plugin.getLogger().info("Language file updated (" + currentLanguage + "): inserted "
+                        + updateResult.insertedPaths + " missing path(s). Backup: " + updateResult.backupFileName);
+            }
+        } else if (!langFile.exists()) {
+            plugin.getLogger().warning("Language resource not found: " + resourcePath + ", file does not exist.");
+        }
+
+        langConfig = YamlConfiguration.loadConfiguration(langFile);
+
+        InputStream defaultLangStream = plugin.getResource(resourcePath);
+        if (defaultLangStream != null) {
+            try (InputStreamReader reader = new InputStreamReader(defaultLangStream, StandardCharsets.UTF_8)) {
+                YamlConfiguration defaultLang = YamlConfiguration.loadConfiguration(reader);
+                langConfig.setDefaults(defaultLang);
+            } catch (IOException ex) {
+                plugin.getLogger().warning("Failed to load default language resource: " + ex.getMessage());
+            }
+        }
+    }
+
+    public void setLanguage(String language) {
+        this.currentLanguage = language;
+        loadLanguage();
+    }
+
+    public String getCurrentLanguage() {
+        return currentLanguage;
+    }
+
+    private String translateColorCodes(String message) {
+        return ChatColor.translateAlternateColorCodes('&', message);
+    }
+
+    public String getMessage(String path) {
+        return translateColorCodes(langConfig.getString("messages." + path, "Missing message: " + path));
+    }
+
+    public String getMessage(String path, String defaultValue) {
+        return translateColorCodes(langConfig.getString("messages." + path, defaultValue));
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/Main.java

@@ -1,23 +1,46 @@
 package com.ddaodan.MineChatGPT;
 
 import org.bukkit.plugin.java.JavaPlugin;
+import org.bstats.bukkit.Metrics;
 
 import java.util.Objects;
 
 public final class Main extends JavaPlugin {
     private ConfigManager configManager;
+    private com.ddaodan.MineChatGPT.service.UserSessionManager sessionManager;
+    private com.ddaodan.MineChatGPT.service.ApiService apiService;
+    private com.ddaodan.MineChatGPT.service.RequestCoordinator requestCoordinator;
     private CommandHandler commandHandler;
+    private MineChatGPTTabCompleter tabCompleter;
 
     @Override
     public void onEnable() {
         saveDefaultConfig();
         configManager = new ConfigManager(this);
-        commandHandler = new CommandHandler(this, configManager);
+        sessionManager = new com.ddaodan.MineChatGPT.service.UserSessionManager(configManager);
+        apiService = new com.ddaodan.MineChatGPT.service.ApiService(this, configManager);
+        requestCoordinator = new com.ddaodan.MineChatGPT.service.RequestCoordinator(this, configManager, apiService, sessionManager);
+        requestCoordinator.start();
+        commandHandler = new CommandHandler(configManager, sessionManager, requestCoordinator);
+        tabCompleter = new MineChatGPTTabCompleter(configManager);
         Objects.requireNonNull(getCommand("chatgpt")).setExecutor(commandHandler);
+        Objects.requireNonNull(getCommand("chatgpt")).setTabCompleter(tabCompleter);
+        if (configManager.isDebugMode()) {
+            getLogger().info( "DEBUG MODE IS TRUE!!!!!");
+        }
+        // Initialize bStats
+        int pluginId = 22635;
+        new Metrics(this, pluginId);
     }
 
     @Override
     public void onDisable() {
+        if (requestCoordinator != null) {
+            requestCoordinator.stop();
+        }
+        if (apiService != null) {
+            apiService.shutdown();
+        }
         saveConfig();
     }
 }

src/main/java/com/ddaodan/MineChatGPT/MineChatGPTTabCompleter.java

@@ -0,0 +1,46 @@
+package com.ddaodan.MineChatGPT;
+
+import org.bukkit.command.Command;
+import org.bukkit.command.CommandSender;
+import org.bukkit.command.TabCompleter;
+
+import java.util.ArrayList;
+import java.util.List;
+public class MineChatGPTTabCompleter implements TabCompleter {
+
+    private final ConfigManager configManager;
+
+    public MineChatGPTTabCompleter(ConfigManager configManager) {
+        this.configManager = configManager;
+    }
+
+    @Override
+    public List<String> onTabComplete(CommandSender sender, Command command, String alias, String[] args) {
+        List<String> completions = new ArrayList<>();
+
+        if (args.length == 1) {
+            // 补全子命令
+            completions.add("reload");
+            completions.add("model");
+            completions.add("modellist");
+            completions.add("context");
+            completions.add("clear");
+            completions.add("character");
+            completions.add("stats");
+        } else if (args.length == 2) {
+            String subCommand = args[0];
+            if (subCommand.equalsIgnoreCase("model")) {
+                completions.addAll(configManager.getModels());
+            } else if (subCommand.equalsIgnoreCase("character")) {
+                completions.addAll(configManager.getCharacters().keySet());
+            } else if (subCommand.equalsIgnoreCase("stats")) {
+                completions.add("reset");
+            }
+        }
+
+        // 过滤补全列表以匹配输入
+        completions.removeIf(completion -> !completion.toLowerCase().startsWith(args[args.length - 1].toLowerCase()));
+
+        return completions;
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/service/ApiService.java

@@ -0,0 +1,201 @@
+package com.ddaodan.MineChatGPT.service;
+
+import com.ddaodan.MineChatGPT.ConfigManager;
+import com.ddaodan.MineChatGPT.Main;
+import jodd.http.HttpRequest;
+import jodd.http.HttpResponse;
+import org.json.JSONArray;
+import org.json.JSONObject;
+
+import java.util.concurrent.CompletableFuture;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
+import java.util.concurrent.ThreadFactory;
+import java.util.concurrent.atomic.AtomicInteger;
+import java.util.logging.Level;
+import java.util.logging.Logger;
+
+/**
+ * API服务类：负责与 OpenAI 兼容接口通信（/chat/completions）
+ */
+public class ApiService {
+    private static final Logger logger = Logger.getLogger(ApiService.class.getName());
+
+    private final Main plugin;
+    private final ConfigManager configManager;
+    private final ExecutorService executor;
+
+    public ApiService(Main plugin, ConfigManager configManager) {
+        this.plugin = plugin;
+        this.configManager = configManager;
+        this.executor = Executors.newFixedThreadPool(
+                configManager.getApiThreadPoolSize(),
+                newNamedThreadFactory("minechatgpt-api-")
+        );
+    }
+
+    public void shutdown() {
+        executor.shutdownNow();
+    }
+
+    public CompletableFuture<ChatCompletionResult> createChatCompletion(String model, JSONArray messages) {
+        if (model == null || model.trim().isEmpty()) {
+            return CompletableFuture.completedFuture(ChatCompletionResult.error("Missing model", null));
+        }
+        if (messages == null) {
+            messages = new JSONArray();
+        }
+
+        String apiKey = configManager.getApiKey();
+        if (apiKey == null || apiKey.trim().isEmpty()) {
+            return CompletableFuture.completedFuture(ChatCompletionResult.error(configManager.getNoApiKeyMessage(), null));
+        }
+
+        JSONObject payload = new JSONObject();
+        payload.put("model", model);
+        payload.put("messages", messages);
+
+        if (configManager.isDebugMode()) {
+            logger.info("Built request: " + payload);
+        }
+
+        String baseUrl = normalizeBaseUrl(configManager.getBaseUrl());
+        HttpRequest request = HttpRequest
+                .post(baseUrl + "/chat/completions")
+                .header("Content-Type", "application/json; charset=UTF-8")
+                .header("Accept", "application/json")
+                .header("Authorization", "Bearer " + apiKey)
+                .header("User-Agent", "MineChatGPT/" + plugin.getDescription().getVersion())
+                .connectionTimeout(configManager.getApiConnectTimeoutMs())
+                .timeout(configManager.getApiTimeoutMs())
+                .bodyText(payload.toString());
+
+        if (configManager.isDebugMode()) {
+            logger.info("Sending request: " + request);
+        }
+
+        return CompletableFuture
+                .supplyAsync(request::send, executor)
+                .thenApply(response -> parseResponse(response))
+                .exceptionally(e -> {
+                    logger.log(Level.SEVERE, "Exception occurred while processing request: " + e.getMessage(), e);
+                    return ChatCompletionResult.error(configManager.getChatGPTErrorMessage(), null);
+                });
+    }
+
+    private ChatCompletionResult parseResponse(HttpResponse response) {
+        if (configManager.isDebugMode()) {
+            logger.info("Received response: " + response);
+        }
+
+        int status = response.statusCode();
+        String body = response.bodyText();
+
+        if (status == 200) {
+            try {
+                JSONObject json = new JSONObject(body);
+                String answer = json
+                        .getJSONArray("choices")
+                        .getJSONObject(0)
+                        .getJSONObject("message")
+                        .getString("content");
+
+                Usage usage = null;
+                if (json.has("usage")) {
+                    JSONObject usageJson = json.getJSONObject("usage");
+                    long prompt = usageJson.optLong("prompt_tokens", -1);
+                    long completion = usageJson.optLong("completion_tokens", -1);
+                    long total = usageJson.optLong("total_tokens", -1);
+                    if (prompt >= 0 && completion >= 0 && total >= 0) {
+                        usage = new Usage(prompt, completion, total);
+                    }
+                }
+
+                return ChatCompletionResult.success(answer, usage);
+            } catch (Exception e) {
+                logger.log(Level.SEVERE, "Failed to parse response: " + e.getMessage(), e);
+                return ChatCompletionResult.error(configManager.getChatGPTErrorMessage(), body);
+            }
+        }
+
+        String debugDetails = tryExtractOpenAiErrorDetails(status, body);
+        logger.log(Level.SEVERE, "Failed to get response (HTTP " + status + "): " + body);
+        return ChatCompletionResult.error(configManager.getChatGPTErrorMessage(), debugDetails);
+    }
+
+    private static String tryExtractOpenAiErrorDetails(int status, String responseBody) {
+        try {
+            JSONObject errorJson = new JSONObject(responseBody);
+            if (!errorJson.has("error")) {
+                return null;
+            }
+            JSONObject err = errorJson.getJSONObject("error");
+            String message = err.optString("message", "");
+            String code = err.optString("code", "");
+            if (message.isEmpty()) {
+                return null;
+            }
+            return "[MineChatGPT] HTTP " + status + (code.isEmpty() ? "" : " (" + code + ")") + ": " + message;
+        } catch (Exception ignored) {
+            return null;
+        }
+    }
+
+    private static String normalizeBaseUrl(String baseUrl) {
+        if (baseUrl == null) {
+            return "https://api.openai.com/v1";
+        }
+        String trimmed = baseUrl.trim();
+        if (trimmed.endsWith("/")) {
+            return trimmed.substring(0, trimmed.length() - 1);
+        }
+        return trimmed;
+    }
+
+    private static ThreadFactory newNamedThreadFactory(String prefix) {
+        AtomicInteger counter = new AtomicInteger(1);
+        return runnable -> {
+            Thread thread = new Thread(runnable, prefix + counter.getAndIncrement());
+            thread.setDaemon(true);
+            return thread;
+        };
+    }
+
+    public static final class Usage {
+        public final long promptTokens;
+        public final long completionTokens;
+        public final long totalTokens;
+
+        public Usage(long promptTokens, long completionTokens, long totalTokens) {
+            this.promptTokens = promptTokens;
+            this.completionTokens = completionTokens;
+            this.totalTokens = totalTokens;
+        }
+    }
+
+    public static final class ChatCompletionResult {
+        public final String answer;
+        public final Usage usage;
+        public final String errorMessage;
+        public final String debugDetails;
+
+        private ChatCompletionResult(String answer, Usage usage, String errorMessage, String debugDetails) {
+            this.answer = answer;
+            this.usage = usage;
+            this.errorMessage = errorMessage;
+            this.debugDetails = debugDetails;
+        }
+
+        public static ChatCompletionResult success(String answer, Usage usage) {
+            return new ChatCompletionResult(answer, usage, null, null);
+        }
+
+        public static ChatCompletionResult error(String errorMessage, String debugDetails) {
+            return new ChatCompletionResult(null, null, errorMessage, debugDetails);
+        }
+
+        public boolean isSuccess() {
+            return errorMessage == null;
+        }
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/service/ApproxCharTokenEstimator.java

@@ -0,0 +1,13 @@
+package com.ddaodan.MineChatGPT.service;
+
+public class ApproxCharTokenEstimator implements TokenEstimator {
+    @Override
+    public int estimateTextTokens(String text) {
+        if (text == null || text.isEmpty()) {
+            return 0;
+        }
+        int chars = text.length();
+        return (int) Math.ceil(chars / 4.0);
+    }
+}
+

src/main/java/com/ddaodan/MineChatGPT/service/CommandService.java

@@ -0,0 +1,218 @@
+package com.ddaodan.MineChatGPT.service;
+
+import com.ddaodan.MineChatGPT.ConfigManager;
+import com.ddaodan.MineChatGPT.ConversationContext;
+import org.bukkit.command.CommandSender;
+
+import java.util.List;
+import java.util.Map;
+
+/**
+ * 命令处理服务类，负责处理各种命令的业务逻辑
+ */
+public class CommandService {
+    private final ConfigManager configManager;
+    private final RequestCoordinator requestCoordinator;
+    private final UserSessionManager sessionManager;
+
+    public CommandService(ConfigManager configManager, RequestCoordinator requestCoordinator, UserSessionManager sessionManager) {
+        this.configManager = configManager;
+        this.requestCoordinator = requestCoordinator;
+        this.sessionManager = sessionManager;
+    }
+
+    /**
+     * 处理重载配置命令
+     *
+     * @param sender 命令发送者
+     * @return 是否成功处理
+     */
+    public boolean handleReloadCommand(CommandSender sender) {
+        if (!sender.hasPermission("minechatgpt.reload")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.reload"));
+            return true;
+        }
+        configManager.reloadConfig();
+        requestCoordinator.reloadFromConfig();
+        sender.sendMessage(configManager.getReloadMessage());
+        return true;
+    }
+
+    /**
+     * 处理模型切换命令
+     *
+     * @param sender 命令发送者
+     * @param args 命令参数
+     * @return 是否成功处理
+     */
+    public boolean handleModelCommand(CommandSender sender, String[] args) {
+        if (!sender.hasPermission("minechatgpt.model")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.model"));
+            return true;
+        }
+        if (args.length < 2) {
+            String currentModel = configManager.getCurrentModel();
+            sender.sendMessage(configManager.getCurrentModelInfoMessage().replace("%s", currentModel));
+            return true;
+        }
+        String model = args[1];
+        List<String> models = configManager.getModels();
+        if (models.contains(model)) {
+            configManager.setCurrentModel(model);
+            sender.sendMessage(configManager.getModelSwitchMessage().replace("%s", model));
+        } else {
+            sender.sendMessage(configManager.getInvalidModelMessage());
+        }
+        return true;
+    }
+
+    /**
+     * 处理模型列表命令
+     *
+     * @param sender 命令发送者
+     * @return 是否成功处理
+     */
+    public boolean handleModelListCommand(CommandSender sender) {
+        if (!sender.hasPermission("minechatgpt.modellist")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.modellist"));
+            return true;
+        }
+        List<String> models = configManager.getModels();
+        sender.sendMessage(configManager.getAvailableModelsMessage());
+        for (String model : models) {
+            sender.sendMessage("- " + model);
+        }
+        return true;
+    }
+
+    /**
+     * 处理上下文切换命令
+     *
+     * @param sender 命令发送者
+     * @param userId 用户ID
+     * @return 是否成功处理
+     */
+    public boolean handleContextCommand(CommandSender sender, String userId) {
+        boolean contextEnabled = !sessionManager.isContextEnabled(userId);
+        sessionManager.setContextEnabled(userId, contextEnabled);
+        String status = contextEnabled ? configManager.getContextToggleEnabledMessage() : configManager.getContextToggleDisabledMessage();
+        sender.sendMessage(configManager.getContextToggleMessage().replace("%s", status));
+        return true;
+    }
+
+    /**
+     * 处理清除历史记录命令
+     *
+     * @param sender 命令发送者
+     * @param userId 用户ID
+     * @return 是否成功处理
+     */
+    public boolean handleClearCommand(CommandSender sender, String userId) {
+        if (!sender.hasPermission("minechatgpt.clear")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.clear"));
+            return true;
+        }
+        sessionManager.clearConversationHistory(userId);
+        sender.sendMessage(configManager.getClearMessage());
+        return true;
+    }
+
+    /**
+     * 处理角色切换命令
+     *
+     * @param sender 命令发送者
+     * @param args 命令参数
+     * @param userId 用户ID
+     * @return 是否成功处理
+     */
+    public boolean handleCharacterCommand(CommandSender sender, String[] args, String userId) {
+        if (!sender.hasPermission("minechatgpt.character")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.character"));
+            return true;
+        }
+        Map<String, String> characters = configManager.getCharacters();
+        if (args.length < 2) {
+            sender.sendMessage(configManager.getAvailableCharactersMessage());
+            for (String character : characters.keySet()) {
+                sender.sendMessage("- " + character);
+            }
+            return true;
+        }
+        String character = args[1];
+        if (characters.containsKey(character)) {
+            sessionManager.setCurrentCharacter(userId, character);
+            sender.sendMessage(configManager.getCharacterSwitchedMessage().replace("%s", character));
+        } else {
+            sender.sendMessage(configManager.getInvalidCharacterMessage());
+        }
+        return true;
+    }
+
+    /**
+     * 处理向ChatGPT提问的命令
+     *
+     * @param sender 命令发送者
+     * @param args 命令参数
+     * @param userId 用户ID
+     * @return 是否成功处理
+     */
+    public boolean handleAskCommand(CommandSender sender, String[] args, String userId) {
+        if (!sender.hasPermission("minechatgpt.use")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.use"));
+            return true;
+        }
+        String question = String.join(" ", args);
+        
+        sender.sendMessage(configManager.getQuestionMessage().replace("%s", question));
+        requestCoordinator.submitAsk(sender, question, userId);
+        return true;
+    }
+
+    public boolean handleStatsCommand(CommandSender sender, String[] args, String userId) {
+        if (!sender.hasPermission("minechatgpt.stats")) {
+            sender.sendMessage(configManager.getNoPermissionMessage().replace("%s", "minechatgpt.stats"));
+            return true;
+        }
+
+        if (args.length >= 2 && "reset".equalsIgnoreCase(args[1])) {
+            requestCoordinator.resetUsageTracker();
+            sender.sendMessage(configManager.getStatsResetMessage());
+            return true;
+        }
+
+        UsageTracker.Snapshot global = requestCoordinator.getUsageTracker().getGlobal();
+        UsageTracker.Snapshot user = requestCoordinator.getUsageTracker().getUser(userId);
+
+        sender.sendMessage(configManager.getStatsHeaderMessage());
+        sender.sendMessage(configManager.getStatsGlobalMessage()
+                .replaceFirst("%s", String.valueOf(global.totalTokens))
+                .replaceFirst("%s", String.valueOf(global.promptTokens))
+                .replaceFirst("%s", String.valueOf(global.completionTokens))
+                .replaceFirst("%s", String.valueOf(global.requests)));
+
+        sender.sendMessage(configManager.getStatsUserMessage()
+                .replaceFirst("%s", String.valueOf(user.totalTokens))
+                .replaceFirst("%s", String.valueOf(user.promptTokens))
+                .replaceFirst("%s", String.valueOf(user.completionTokens))
+                .replaceFirst("%s", String.valueOf(user.requests)));
+
+        return true;
+    }
+
+    /**
+     * 发送帮助信息
+     *
+     * @param sender 命令发送者
+     */
+    public void sendHelpMessage(CommandSender sender) {
+        sender.sendMessage(configManager.getHelpMessage());
+        sender.sendMessage(configManager.getHelpAskMessage());
+        sender.sendMessage(configManager.getHelpReloadMessage());
+        sender.sendMessage(configManager.getHelpModelMessage());
+        sender.sendMessage(configManager.getHelpModelListMessage());
+        sender.sendMessage(configManager.getHelpContextMessage());
+        sender.sendMessage(configManager.getHelpClearMessage());
+        sender.sendMessage(configManager.getHelpCharacterMessage());
+        sender.sendMessage(configManager.getHelpStatsMessage());
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/service/RateLimiter.java

@@ -0,0 +1,110 @@
+package com.ddaodan.MineChatGPT.service;
+
+import com.ddaodan.MineChatGPT.ConfigManager;
+
+import java.util.Map;
+import java.util.concurrent.ConcurrentHashMap;
+
+public class RateLimiter {
+    public static final class Decision {
+        public final boolean allowed;
+        public final String reason;
+        public final long retryAfterMs;
+
+        private Decision(boolean allowed, String reason, long retryAfterMs) {
+            this.allowed = allowed;
+            this.reason = reason;
+            this.retryAfterMs = retryAfterMs;
+        }
+
+        public static Decision allow() {
+            return new Decision(true, null, 0);
+        }
+
+        public static Decision deny(String reason, long retryAfterMs) {
+            return new Decision(false, reason, retryAfterMs);
+        }
+    }
+
+    private static final class UserBuckets {
+        private final TokenBucket requests;
+        private final TokenBucket tokens;
+
+        private UserBuckets(TokenBucket requests, TokenBucket tokens) {
+            this.requests = requests;
+            this.tokens = tokens;
+        }
+    }
+
+    private final ConfigManager configManager;
+    private final Map<String, UserBuckets> perUser = new ConcurrentHashMap<>();
+    private final Map<String, Long> nextAllowedAt = new ConcurrentHashMap<>();
+
+    private volatile TokenBucket globalRequests;
+    private volatile TokenBucket globalTokens;
+
+    public RateLimiter(ConfigManager configManager) {
+        this.configManager = configManager;
+        long now = System.currentTimeMillis();
+        this.globalRequests = new TokenBucket(
+                configManager.getGlobalBurstRequests(),
+                configManager.getGlobalRequestsPerMinute() / 60.0,
+                now
+        );
+        this.globalTokens = new TokenBucket(
+                configManager.getGlobalBurstTokens(),
+                configManager.getGlobalTokensPerMinute() / 60.0,
+                now
+        );
+    }
+
+    public Decision tryAcquire(String userId, int estimatedTotalTokens, long nowMillis) {
+        if (!configManager.isRateLimitEnabled()) {
+            return Decision.allow();
+        }
+
+        long cooldownMs = configManager.getRateLimitCooldownMs();
+        Long next = nextAllowedAt.get(userId);
+        if (next != null && nowMillis < next) {
+            return Decision.deny("cooldown", next - nowMillis);
+        }
+
+        String mode = configManager.getRateLimitMode();
+        boolean limitRequests = "requests".equalsIgnoreCase(mode) || "both".equalsIgnoreCase(mode);
+        boolean limitTokens = "tokens".equalsIgnoreCase(mode) || "both".equalsIgnoreCase(mode);
+
+        UserBuckets buckets = perUser.computeIfAbsent(userId, id -> {
+            TokenBucket req = new TokenBucket(
+                    configManager.getPerUserBurstRequests(),
+                    configManager.getPerUserRequestsPerMinute() / 60.0,
+                    nowMillis
+            );
+            TokenBucket tok = new TokenBucket(
+                    configManager.getPerUserBurstTokens(),
+                    configManager.getPerUserTokensPerMinute() / 60.0,
+                    nowMillis
+            );
+            return new UserBuckets(req, tok);
+        });
+
+        if (limitRequests) {
+            if (!globalRequests.tryConsume(1.0, nowMillis) || !buckets.requests.tryConsume(1.0, nowMillis)) {
+                return Decision.deny("rate_limited", 0);
+            }
+        }
+
+        if (limitTokens) {
+            double cost = Math.max(0, estimatedTotalTokens);
+            if (!globalTokens.tryConsume(cost, nowMillis) || !buckets.tokens.tryConsume(cost, nowMillis)) {
+                return Decision.deny("rate_limited", 0);
+            }
+        }
+
+        if (cooldownMs > 0) {
+            nextAllowedAt.put(userId, nowMillis + cooldownMs);
+        }
+
+        return Decision.allow();
+    }
+}
+

src/main/java/com/ddaodan/MineChatGPT/service/RequestCoordinator.java

@@ -0,0 +1,404 @@
+package com.ddaodan.MineChatGPT.service;
+
+import com.ddaodan.MineChatGPT.ConfigManager;
+import com.ddaodan.MineChatGPT.ConversationContext;
+import com.ddaodan.MineChatGPT.Main;
+import org.bukkit.Bukkit;
+import org.bukkit.command.CommandSender;
+import org.bukkit.entity.Player;
+import org.json.JSONArray;
+import org.json.JSONObject;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.concurrent.CompletableFuture;
+import java.util.concurrent.atomic.AtomicInteger;
+import java.util.logging.Level;
+
+public class RequestCoordinator {
+    private static final String SUMMARY_SYSTEM_PROMPT =
+            "You are a summarization assistant. Summarize the conversation for future context. " +
+            "Keep it concise, preserve important facts, user preferences, goals, and constraints. " +
+            "Return ONLY the updated summary text.";
+
+    private final Main plugin;
+    private final ConfigManager configManager;
+    private final ApiService apiService;
+    private final UserSessionManager sessionManager;
+    private volatile RateLimiter rateLimiter;
+    private final AtomicInteger inFlight = new AtomicInteger(0);
+
+    private volatile UsageTracker usageTracker;
+    private volatile TokenEstimator tokenEstimator;
+    private volatile RequestQueue<RequestJob> queue;
+    private volatile int dispatchTaskId = -1;
+
+    public RequestCoordinator(Main plugin, ConfigManager configManager, ApiService apiService, UserSessionManager sessionManager) {
+        this.plugin = plugin;
+        this.configManager = configManager;
+        this.apiService = apiService;
+        this.sessionManager = sessionManager;
+        this.usageTracker = new UsageTracker();
+        reloadFromConfig();
+    }
+
+    public void reloadFromConfig() {
+        this.tokenEstimator = createTokenEstimator(configManager.getRateLimitTokenEstimator());
+        this.queue = new RequestQueue<>(configManager.getQueueMaxSize(), configManager.getQueueMaxPerUser());
+        this.rateLimiter = new RateLimiter(configManager);
+    }
+
+    public UsageTracker getUsageTracker() {
+        return usageTracker;
+    }
+
+    public void resetUsageTracker() {
+        this.usageTracker = new UsageTracker();
+    }
+
+    public void start() {
+        if (dispatchTaskId != -1) {
+            return;
+        }
+        dispatchTaskId = Bukkit.getScheduler().runTaskTimer(plugin, () -> {
+            if (!configManager.isQueueEnabled()) {
+                return;
+            }
+            int maxInFlight = configManager.getQueueMaxInFlight();
+            int perTick = configManager.getQueueDispatchPerTick();
+            for (int i = 0; i < perTick; i++) {
+                if (inFlight.get() >= maxInFlight) {
+                    return;
+                }
+                RequestJob job = queue.poll();
+                if (job == null) {
+                    return;
+                }
+                dispatch(job);
+            }
+        }, 1L, 1L).getTaskId();
+    }
+
+    public void stop() {
+        if (dispatchTaskId != -1) {
+            Bukkit.getScheduler().cancelTask(dispatchTaskId);
+            dispatchTaskId = -1;
+        }
+    }
+
+    public void submitAsk(CommandSender sender, String question, String userId) {
+        boolean contextEnabled = sessionManager.isContextEnabled(userId);
+        ConversationContext context = sessionManager.getConversationContext(userId);
+
+        int estimatedTokensForLimit = estimateTokensForLimit(userId, question, contextEnabled, context);
+        long now = System.currentTimeMillis();
+        RateLimiter.Decision decision = rateLimiter.tryAcquire(userId, estimatedTokensForLimit, now);
+        if (!decision.allowed) {
+            if ("cooldown".equals(decision.reason)) {
+                sender.sendMessage(configManager.getCooldownMessage().replace("%s", String.valueOf(decision.retryAfterMs)));
+            } else {
+                sender.sendMessage(configManager.getRateLimitedMessage());
+            }
+            return;
+        }
+
+        RequestJob job = new RequestJob(sender, question, userId, contextEnabled, estimatedTokensForLimit);
+
+        if (!configManager.isQueueEnabled()) {
+            int maxInFlight = configManager.getQueueMaxInFlight();
+            if (inFlight.get() >= maxInFlight) {
+                sender.sendMessage(configManager.getQueueFullMessage());
+                return;
+            }
+            dispatch(job);
+            return;
+        }
+
+        RequestQueue.EnqueueResult result = queue.tryEnqueue(userId, job);
+        if (!result.accepted) {
+            if ("queue_full_user".equals(result.reason)) {
+                sender.sendMessage(configManager.getQueueFullUserMessage());
+            } else {
+                sender.sendMessage(configManager.getQueueFullMessage());
+            }
+            return;
+        }
+
+        sender.sendMessage(configManager.getQueuedMessage().replace("%s", String.valueOf(result.position)));
+    }
+
+    private void dispatch(RequestJob job) {
+        inFlight.incrementAndGet();
+        process(job)
+                .whenComplete((ignored, e) -> {
+                    if (e != null) {
+                        plugin.getLogger().log(Level.SEVERE, "Job failed: " + e.getMessage(), e);
+                    }
+                    inFlight.decrementAndGet();
+                });
+    }
+
+    private CompletableFuture<Void> process(RequestJob job) {
+        if (job.sender instanceof Player) {
+            Player player = (Player) job.sender;
+            if (!player.isOnline()) {
+                return CompletableFuture.completedFuture(null);
+            }
+        }
+
+        ConversationContext context = sessionManager.getConversationContext(job.userId);
+        String characterName = sessionManager.getCurrentCharacter(job.userId);
+        String characterPrompt = configManager.getCharacters().get(characterName);
+        String summary = sessionManager.getSummary(job.userId);
+
+        CompletableFuture<Void> maybeSummarize = ensureContextWithinBudget(job.userId, job.contextEnabled, characterPrompt, summary, context);
+
+        return maybeSummarize.thenCompose(v -> {
+            String updatedSummary = sessionManager.getSummary(job.userId);
+            JSONArray messages = buildMessages(job.contextEnabled, characterPrompt, updatedSummary, context, job.question);
+            String model = configManager.getCurrentModel();
+            return apiService.createChatCompletion(model, messages)
+                    .thenAccept(result -> handleCompletion(job, characterName, result));
+        });
+    }
+
+    private void handleCompletion(RequestJob job, String characterName, ApiService.ChatCompletionResult result) {
+        Bukkit.getScheduler().runTask(plugin, () -> {
+            if (!plugin.isEnabled()) {
+                return;
+            }
+            if (job.sender instanceof Player) {
+                Player player = (Player) job.sender;
+                if (!player.isOnline()) {
+                    return;
+                }
+            }
+
+            if (!result.isSuccess()) {
+                job.sender.sendMessage(result.errorMessage);
+                if (configManager.isDebugMode() && result.debugDetails != null && !result.debugDetails.isEmpty()) {
+                    job.sender.sendMessage(result.debugDetails);
+                }
+                return;
+            }
+
+            String formatted = formatTwoPlaceholders(configManager.getChatGPTResponseMessage(), characterName, result.answer);
+            sendMultiline(job.sender, formatted);
+
+            if (job.contextEnabled) {
+                ConversationContext ctx = sessionManager.getConversationContext(job.userId);
+                ctx.addUserMessage(job.question);
+                ctx.addAssistantMessage(result.answer);
+            }
+
+            long promptTokens = 0;
+            long completionTokens = 0;
+            long totalTokens = 0;
+            if (result.usage != null) {
+                promptTokens = result.usage.promptTokens;
+                completionTokens = result.usage.completionTokens;
+                totalTokens = result.usage.totalTokens;
+            } else {
+                // fallback estimation
+                int estimatedPrompt = estimatePromptTokens(job.userId, job.question, job.contextEnabled);
+                int estimatedCompletion = tokenEstimator.estimateTextTokens(result.answer);
+                promptTokens = Math.max(0, estimatedPrompt);
+                completionTokens = Math.max(0, estimatedCompletion);
+                totalTokens = promptTokens + completionTokens;
+            }
+            usageTracker.record(job.userId, promptTokens, completionTokens, totalTokens);
+        });
+    }
+
+    private CompletableFuture<Void> ensureContextWithinBudget(String userId, boolean contextEnabled, String characterPrompt, String summary, ConversationContext context) {
+        if (!contextEnabled) {
+            return CompletableFuture.completedFuture(null);
+        }
+
+        if (!configManager.isSummarizationEnabled()) {
+            trimContextToBudget(characterPrompt, summary, context);
+            return CompletableFuture.completedFuture(null);
+        }
+
+        int trigger = configManager.getSummarizationTriggerTokens();
+        int estimated = estimateContextTokens(characterPrompt, summary, context);
+        if (estimated <= trigger) {
+            trimContextToBudget(characterPrompt, summary, context);
+            return CompletableFuture.completedFuture(null);
+        }
+
+        int keepLast = configManager.getSummarizationKeepLastMessages();
+        List<ConversationContext.Message> history = context.getMessages();
+        if (history.size() <= keepLast) {
+            trimContextToBudget(characterPrompt, summary, context);
+            return CompletableFuture.completedFuture(null);
+        }
+
+        List<ConversationContext.Message> toSummarize = history.subList(0, history.size() - keepLast);
+        List<ConversationContext.Message> toKeep = history.subList(history.size() - keepLast, history.size());
+
+        JSONArray summarizeMessages = new JSONArray();
+        summarizeMessages.put(new JSONObject().put("role", "system").put("content", SUMMARY_SYSTEM_PROMPT));
+
+        StringBuilder sb = new StringBuilder();
+        String existing = sessionManager.getSummary(userId);
+        if (existing != null && !existing.trim().isEmpty()) {
+            sb.append("Current summary:\n").append(existing.trim()).append("\n\n");
+        }
+        sb.append("Conversation to summarize:\n");
+        for (ConversationContext.Message msg : toSummarize) {
+            sb.append(msg.getRole()).append(": ").append(msg.getContent()).append("\n");
+        }
+        summarizeMessages.put(new JSONObject().put("role", "user").put("content", sb.toString()));
+
+        String model = configManager.getSummarizationModel();
+        return apiService.createChatCompletion(model, summarizeMessages)
+                .thenCompose(result -> runSync(() -> {
+                    if (!result.isSuccess()) {
+                        return;
+                    }
+                    String newSummary = result.answer == null ? "" : result.answer.trim();
+                    sessionManager.setSummary(userId, newSummary);
+                    context.setMessages(new ArrayList<>(toKeep));
+                    trimContextToBudget(characterPrompt, newSummary, context);
+
+                    if (result.usage != null) {
+                        usageTracker.record(userId, result.usage.promptTokens, result.usage.completionTokens, result.usage.totalTokens);
+                    }
+                }));
+    }
+
+    private CompletableFuture<Void> runSync(Runnable runnable) {
+        CompletableFuture<Void> future = new CompletableFuture<>();
+        Bukkit.getScheduler().runTask(plugin, () -> {
+            try {
+                runnable.run();
+                future.complete(null);
+            } catch (Exception e) {
+                future.completeExceptionally(e);
+            }
+        });
+        return future;
+    }
+
+    private void trimContextToBudget(String characterPrompt, String summary, ConversationContext context) {
+        int budget = configManager.getMaxContextTokens();
+        if (budget <= 0) {
+            return;
+        }
+        while (estimateContextTokens(characterPrompt, summary, context) > budget) {
+            List<ConversationContext.Message> history = context.getMessages();
+            if (history.isEmpty()) {
+                return;
+            }
+            context.removeOldestMessage();
+        }
+    }
+
+    private int estimateContextTokens(String characterPrompt, String summary, ConversationContext context) {
+        int total = 0;
+        if (characterPrompt != null && !characterPrompt.isEmpty()) {
+            total += tokenEstimator.estimateMessageTokens("system", characterPrompt);
+        }
+        if (summary != null && !summary.trim().isEmpty()) {
+            total += tokenEstimator.estimateMessageTokens("system", "Conversation summary:\n" + summary.trim());
+        }
+        for (ConversationContext.Message msg : context.getMessages()) {
+            total += tokenEstimator.estimateMessageTokens(msg.getRole(), msg.getContent());
+        }
+        return total;
+    }
+
+    private int estimatePromptTokens(String userId, String question, boolean contextEnabled) {
+        ConversationContext context = sessionManager.getConversationContext(userId);
+        String characterName = sessionManager.getCurrentCharacter(userId);
+        String characterPrompt = configManager.getCharacters().get(characterName);
+        String summary = sessionManager.getSummary(userId);
+
+        int total = 0;
+        if (characterPrompt != null && !characterPrompt.isEmpty()) {
+            total += tokenEstimator.estimateMessageTokens("system", characterPrompt);
+        }
+        if (contextEnabled) {
+            if (summary != null && !summary.trim().isEmpty()) {
+                total += tokenEstimator.estimateMessageTokens("system", "Conversation summary:\n" + summary.trim());
+            }
+            for (ConversationContext.Message msg : context.getMessages()) {
+                total += tokenEstimator.estimateMessageTokens(msg.getRole(), msg.getContent());
+            }
+        }
+        total += tokenEstimator.estimateMessageTokens("user", question);
+        return total;
+    }
+
+    private int estimateTokensForLimit(String userId, String question, boolean contextEnabled, ConversationContext context) {
+        int assumedCompletion = configManager.getRateLimitAssumedCompletionTokens();
+        int reserve = configManager.getReserveCompletionTokens();
+        int completionBudget = Math.max(assumedCompletion, reserve);
+        int prompt = estimatePromptTokens(userId, question, contextEnabled);
+        return prompt + completionBudget;
+    }
+
+    private JSONArray buildMessages(boolean contextEnabled, String characterPrompt, String summary, ConversationContext context, String question) {
+        JSONArray messages = new JSONArray();
+        if (characterPrompt != null && !characterPrompt.isEmpty()) {
+            messages.put(new JSONObject().put("role", "system").put("content", characterPrompt));
+        }
+        if (contextEnabled) {
+            if (summary != null && !summary.trim().isEmpty()) {
+                messages.put(new JSONObject().put("role", "system").put("content", "Conversation summary:\n" + summary.trim()));
+            }
+            for (ConversationContext.Message msg : context.getMessages()) {
+                messages.put(new JSONObject().put("role", msg.getRole()).put("content", msg.getContent()));
+            }
+        }
+        messages.put(new JSONObject().put("role", "user").put("content", question));
+        return messages;
+    }
+
+    private static void sendMultiline(CommandSender sender, String message) {
+        String[] lines = message.split("\\R", -1);
+        for (String line : lines) {
+            if (!line.isEmpty()) {
+                sender.sendMessage(line);
+            }
+        }
+    }
+
+    private static String formatTwoPlaceholders(String template, String first, String second) {
+        String withFirst = replaceFirstLiteral(template, "%s", first);
+        return replaceFirstLiteral(withFirst, "%s", second);
+    }
+
+    private static String replaceFirstLiteral(String template, String token, String replacement) {
+        int index = template.indexOf(token);
+        if (index < 0) {
+            return template;
+        }
+        return template.substring(0, index) + replacement + template.substring(index + token.length());
+    }
+
+    private static TokenEstimator createTokenEstimator(String type) {
+        if ("approx_chars".equalsIgnoreCase(type)) {
+            return new ApproxCharTokenEstimator();
+        }
+        return new ApproxCharTokenEstimator();
+    }
+
+    private static final class RequestJob {
+        private final CommandSender sender;
+        private final String question;
+        private final String userId;
+        private final boolean contextEnabled;
+        @SuppressWarnings("unused")
+        private final int estimatedTokensForLimit;
+
+        private RequestJob(CommandSender sender, String question, String userId, boolean contextEnabled, int estimatedTokensForLimit) {
+            this.sender = sender;
+            this.question = question;
+            this.userId = userId;
+            this.contextEnabled = contextEnabled;
+            this.estimatedTokensForLimit = estimatedTokensForLimit;
+        }
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/service/RequestQueue.java

@@ -0,0 +1,83 @@
+package com.ddaodan.MineChatGPT.service;
+
+import java.util.ArrayDeque;
+import java.util.Deque;
+import java.util.HashMap;
+import java.util.Map;
+
+public class RequestQueue<T> {
+    public static final class EnqueueResult {
+        public final boolean accepted;
+        public final int position;
+        public final String reason;
+
+        private EnqueueResult(boolean accepted, int position, String reason) {
+            this.accepted = accepted;
+            this.position = position;
+            this.reason = reason;
+        }
+
+        public static EnqueueResult accepted(int position) {
+            return new EnqueueResult(true, position, null);
+        }
+
+        public static EnqueueResult rejected(String reason) {
+            return new EnqueueResult(false, -1, reason);
+        }
+    }
+
+    private final int maxSize;
+    private final int maxPerUser;
+    private final Deque<Entry<T>> queue = new ArrayDeque<>();
+    private final Map<String, Integer> perUserCounts = new HashMap<>();
+
+    private static final class Entry<T> {
+        private final String userId;
+        private final T item;
+
+        private Entry(String userId, T item) {
+            this.userId = userId;
+            this.item = item;
+        }
+    }
+
+    public RequestQueue(int maxSize, int maxPerUser) {
+        this.maxSize = Math.max(0, maxSize);
+        this.maxPerUser = Math.max(0, maxPerUser);
+    }
+
+    public synchronized EnqueueResult tryEnqueue(String userId, T item) {
+        if (maxSize > 0 && queue.size() >= maxSize) {
+            return EnqueueResult.rejected("queue_full");
+        }
+        int count = perUserCounts.getOrDefault(userId, 0);
+        if (maxPerUser > 0 && count >= maxPerUser) {
+            return EnqueueResult.rejected("queue_full_user");
+        }
+        queue.addLast(new Entry<>(userId, item));
+        perUserCounts.put(userId, count + 1);
+        return EnqueueResult.accepted(queue.size());
+    }
+
+    public synchronized T poll() {
+        Entry<T> entry = queue.pollFirst();
+        if (entry == null) {
+            return null;
+        }
+        Integer count = perUserCounts.get(entry.userId);
+        if (count != null) {
+            int next = count - 1;
+            if (next <= 0) {
+                perUserCounts.remove(entry.userId);
+            } else {
+                perUserCounts.put(entry.userId, next);
+            }
+        }
+        return entry.item;
+    }
+
+    public synchronized int size() {
+        return queue.size();
+    }
+}
+

src/main/java/com/ddaodan/MineChatGPT/service/TokenBucket.java

@@ -0,0 +1,48 @@
+package com.ddaodan.MineChatGPT.service;
+
+public class TokenBucket {
+    private final double capacity;
+    private final double refillTokensPerMillis;
+    private double tokens;
+    private long lastRefillMillis;
+
+    public TokenBucket(double capacity, double refillTokensPerSecond, long nowMillis) {
+        this.capacity = Math.max(0.0, capacity);
+        this.refillTokensPerMillis = Math.max(0.0, refillTokensPerSecond) / 1000.0;
+        this.tokens = this.capacity;
+        this.lastRefillMillis = nowMillis;
+    }
+
+    public synchronized boolean tryConsume(double amount, long nowMillis) {
+        refill(nowMillis);
+        if (amount <= 0) {
+            return true;
+        }
+        if (tokens >= amount) {
+            tokens -= amount;
+            return true;
+        }
+        return false;
+    }
+
+    public synchronized void refund(double amount, long nowMillis) {
+        refill(nowMillis);
+        if (amount <= 0) {
+            return;
+        }
+        tokens = Math.min(capacity, tokens + amount);
+    }
+
+    private void refill(long nowMillis) {
+        if (nowMillis <= lastRefillMillis) {
+            return;
+        }
+        double elapsed = nowMillis - lastRefillMillis;
+        double add = elapsed * refillTokensPerMillis;
+        if (add > 0) {
+            tokens = Math.min(capacity, tokens + add);
+        }
+        lastRefillMillis = nowMillis;
+    }
+}
+

src/main/java/com/ddaodan/MineChatGPT/service/TokenEstimator.java

@@ -0,0 +1,11 @@
+package com.ddaodan.MineChatGPT.service;
+
+public interface TokenEstimator {
+    int estimateTextTokens(String text);
+
+    default int estimateMessageTokens(String role, String content) {
+        int base = 4; // message overhead (approx)
+        return base + estimateTextTokens(role) + estimateTextTokens(content);
+    }
+}
+

src/main/java/com/ddaodan/MineChatGPT/service/UsageTracker.java

@@ -0,0 +1,72 @@
+package com.ddaodan.MineChatGPT.service;
+
+import java.util.Map;
+import java.util.concurrent.ConcurrentHashMap;
+import java.util.concurrent.atomic.LongAdder;
+
+public class UsageTracker {
+    public static final class Snapshot {
+        public final long requests;
+        public final long promptTokens;
+        public final long completionTokens;
+        public final long totalTokens;
+
+        private Snapshot(long requests, long promptTokens, long completionTokens, long totalTokens) {
+            this.requests = requests;
+            this.promptTokens = promptTokens;
+            this.completionTokens = completionTokens;
+            this.totalTokens = totalTokens;
+        }
+    }
+
+    private static final class Counters {
+        private final LongAdder requests = new LongAdder();
+        private final LongAdder promptTokens = new LongAdder();
+        private final LongAdder completionTokens = new LongAdder();
+        private final LongAdder totalTokens = new LongAdder();
+
+        private Snapshot snapshot() {
+            return new Snapshot(
+                    requests.sum(),
+                    promptTokens.sum(),
+                    completionTokens.sum(),
+                    totalTokens.sum()
+            );
+        }
+    }
+
+    private final Counters global = new Counters();
+    private final Map<String, Counters> perUser = new ConcurrentHashMap<>();
+
+    public void record(String userId, long promptTokens, long completionTokens, long totalTokens) {
+        global.requests.increment();
+        global.promptTokens.add(promptTokens);
+        global.completionTokens.add(completionTokens);
+        global.totalTokens.add(totalTokens);
+
+        Counters user = perUser.computeIfAbsent(userId, k -> new Counters());
+        user.requests.increment();
+        user.promptTokens.add(promptTokens);
+        user.completionTokens.add(completionTokens);
+        user.totalTokens.add(totalTokens);
+    }
+
+    public Snapshot getGlobal() {
+        return global.snapshot();
+    }
+
+    public Snapshot getUser(String userId) {
+        Counters user = perUser.get(userId);
+        if (user == null) {
+            return new Snapshot(0, 0, 0, 0);
+        }
+        return user.snapshot();
+    }
+
+    public void reset() {
+        perUser.clear();
+        // global LongAdder has no reset without recreating
+        // Create a new UsageTracker if you need a hard reset
+    }
+}
+

src/main/java/com/ddaodan/MineChatGPT/service/UserSessionManager.java

@@ -0,0 +1,114 @@
+package com.ddaodan.MineChatGPT.service;
+
+import com.ddaodan.MineChatGPT.ConfigManager;
+import com.ddaodan.MineChatGPT.ConversationContext;
+
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * 用户会话管理类，负责管理用户的会话状态和上下文
+ */
+public class UserSessionManager {
+    private final ConfigManager configManager;
+    private final Map<String, ConversationContext> userContexts;
+    private final Map<String, Boolean> userContextEnabled;
+    private final Map<String, String> userCurrentCharacter;
+    private final Map<String, String> userSummaries;
+
+    public UserSessionManager(ConfigManager configManager) {
+        this.configManager = configManager;
+        this.userContexts = new HashMap<>();
+        this.userContextEnabled = new HashMap<>();
+        this.userCurrentCharacter = new HashMap<>();
+        this.userSummaries = new HashMap<>();
+    }
+
+    /**
+     * 获取用户的对话上下文
+     *
+     * @param userId 用户ID
+     * @return 对话上下文
+     */
+    public ConversationContext getConversationContext(String userId) {
+        if (!userContexts.containsKey(userId)) {
+            userContexts.put(userId, new ConversationContext(configManager.getMaxHistorySize()));
+            userContextEnabled.put(userId, configManager.isContextEnabled());
+        }
+        return userContexts.get(userId);
+    }
+
+    /**
+     * 判断用户是否启用了上下文
+     *
+     * @param userId 用户ID
+     * @return 是否启用上下文
+     */
+    public boolean isContextEnabled(String userId) {
+        if (!userContextEnabled.containsKey(userId)) {
+            userContextEnabled.put(userId, configManager.isContextEnabled());
+        }
+        return userContextEnabled.get(userId);
+    }
+
+    /**
+     * 设置用户的上下文启用状态
+     *
+     * @param userId 用户ID
+     * @param enabled 是否启用
+     */
+    public void setContextEnabled(String userId, boolean enabled) {
+        userContextEnabled.put(userId, enabled);
+    }
+
+    /**
+     * 清除用户的对话历史
+     *
+     * @param userId 用户ID
+     */
+    public void clearConversationHistory(String userId) {
+        ConversationContext context = getConversationContext(userId);
+        context.clearHistory();
+        clearSummary(userId);
+    }
+
+    /**
+     * 获取用户当前使用的角色
+     *
+     * @param userId 用户ID
+     * @return 角色名称
+     */
+    public String getCurrentCharacter(String userId) {
+        if (!userCurrentCharacter.containsKey(userId)) {
+            userCurrentCharacter.put(userId, configManager.getCurrentCharacter(userId));
+        }
+        return userCurrentCharacter.get(userId);
+    }
+
+    /**
+     * 设置用户当前使用的角色
+     *
+     * @param userId 用户ID
+     * @param character 角色名称
+     */
+    public void setCurrentCharacter(String userId, String character) {
+        userCurrentCharacter.put(userId, character);
+        configManager.setCurrentCharacter(userId, character);
+    }
+
+    public String getSummary(String userId) {
+        return userSummaries.getOrDefault(userId, "");
+    }
+
+    public void setSummary(String userId, String summary) {
+        if (summary == null) {
+            userSummaries.remove(userId);
+            return;
+        }
+        userSummaries.put(userId, summary);
+    }
+
+    public void clearSummary(String userId) {
+        userSummaries.remove(userId);
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/util/ChatUtils.java

@@ -0,0 +1,9 @@
+package com.ddaodan.MineChatGPT.util;
+
+import org.bukkit.ChatColor;
+
+public class ChatUtils {
+    public static String translateColorCodes(String message) {
+        return ChatColor.translateAlternateColorCodes('&', message);
+    }
+}

src/main/java/com/ddaodan/MineChatGPT/util/ConfigFileUpdater.java

@@ -0,0 +1,424 @@
+package com.ddaodan.MineChatGPT.util;
+
+import org.bukkit.configuration.file.YamlConfiguration;
+import org.bukkit.plugin.java.JavaPlugin;
+
+import java.io.BufferedReader;
+import java.io.File;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.Reader;
+import java.nio.charset.StandardCharsets;
+import java.nio.file.Files;
+import java.nio.file.StandardCopyOption;
+import java.text.SimpleDateFormat;
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.Date;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Locale;
+import java.util.Map;
+import java.util.Set;
+import java.util.TreeSet;
+
+public final class ConfigFileUpdater {
+    private ConfigFileUpdater() {
+    }
+
+    public static UpdateResult updateIfMissingKeys(JavaPlugin plugin, String resourcePath) {
+        File configFile = new File(plugin.getDataFolder(), resourcePath);
+        if (!configFile.exists()) {
+            plugin.saveResource(resourcePath, false);
+            return UpdateResult.noChange();
+        }
+
+        try {
+            YamlConfiguration currentYaml = YamlConfiguration.loadConfiguration(configFile);
+            YamlConfiguration defaultYaml = loadYamlFromResource(plugin, resourcePath);
+            if (defaultYaml == null) {
+                return UpdateResult.noChange();
+            }
+
+            List<String> defaultLines = readResourceLines(plugin, resourcePath);
+            if (defaultLines.isEmpty()) {
+                return UpdateResult.noChange();
+            }
+            DocumentModel defaultModel = parseDocument(defaultLines);
+
+            List<String> missingTopLevelPaths = collectMissingTopPaths(currentYaml, defaultYaml, defaultModel);
+            if (missingTopLevelPaths.isEmpty()) {
+                return UpdateResult.noChange();
+            }
+
+            List<String> currentLines = Files.readAllLines(configFile.toPath(), StandardCharsets.UTF_8);
+            int inserted = 0;
+            for (String missingPath : missingTopLevelPaths) {
+                Entry source = defaultModel.entriesByPath.get(missingPath);
+                if (source == null) {
+                    continue;
+                }
+
+                List<String> snippet = new ArrayList<>(
+                        defaultLines.subList(source.startLine, source.endLine + 1)
+                );
+                if (snippet.isEmpty()) {
+                    continue;
+                }
+
+                DocumentModel currentModel = parseDocument(currentLines);
+                int insertAt = findInsertIndex(currentLines, currentModel, defaultModel, missingPath);
+                if (insertAt < 0 || insertAt > currentLines.size()) {
+                    insertAt = currentLines.size();
+                }
+                insertSnippetWithSpacing(currentLines, insertAt, snippet);
+                inserted++;
+            }
+
+            if (inserted == 0) {
+                return UpdateResult.noChange();
+            }
+
+            File backup = createBackup(configFile);
+            Files.write(configFile.toPath(), currentLines, StandardCharsets.UTF_8);
+            return UpdateResult.updated(inserted, backup.getName());
+        } catch (Exception ex) {
+            plugin.getLogger().warning("Failed to auto-update config.yml: " + ex.getMessage());
+            return UpdateResult.noChange();
+        }
+    }
+
+    private static YamlConfiguration loadYamlFromResource(JavaPlugin plugin, String resourcePath) {
+        InputStream inputStream = plugin.getResource(resourcePath);
+        if (inputStream == null) {
+            return null;
+        }
+        try (Reader reader = new InputStreamReader(inputStream, StandardCharsets.UTF_8)) {
+            return YamlConfiguration.loadConfiguration(reader);
+        } catch (IOException ex) {
+            plugin.getLogger().warning("Failed to read default config resource: " + ex.getMessage());
+            return null;
+        }
+    }
+
+    private static List<String> collectMissingTopPaths(
+            YamlConfiguration currentYaml,
+            YamlConfiguration defaultYaml,
+            DocumentModel defaultModel
+    ) {
+        Set<String> missing = new TreeSet<String>();
+        for (String key : defaultYaml.getKeys(true)) {
+            if (!currentYaml.contains(key)) {
+                missing.add(key);
+            }
+        }
+        if (missing.isEmpty()) {
+            return Collections.emptyList();
+        }
+
+        List<String> top = new ArrayList<String>();
+        for (Entry entry : defaultModel.entriesInOrder) {
+            String path = entry.path;
+            if (!missing.contains(path)) {
+                continue;
+            }
+            if (!hasAncestor(top, path)) {
+                top.add(path);
+            }
+        }
+        return top;
+    }
+
+    private static boolean hasAncestor(List<String> selected, String path) {
+        int index = path.lastIndexOf('.');
+        while (index > 0) {
+            String parent = path.substring(0, index);
+            if (selected.contains(parent)) {
+                return true;
+            }
+            index = parent.lastIndexOf('.');
+        }
+        return false;
+    }
+
+    private static List<String> readResourceLines(JavaPlugin plugin, String resourcePath) throws IOException {
+        InputStream inputStream = plugin.getResource(resourcePath);
+        if (inputStream == null) {
+            return Collections.emptyList();
+        }
+        List<String> lines = new ArrayList<String>();
+        try (BufferedReader reader = new BufferedReader(new InputStreamReader(inputStream, StandardCharsets.UTF_8))) {
+            String line;
+            while ((line = reader.readLine()) != null) {
+                lines.add(line);
+            }
+        }
+        return lines;
+    }
+
+    private static int findInsertIndex(
+            List<String> currentLines,
+            DocumentModel currentModel,
+            DocumentModel defaultModel,
+            String path
+    ) {
+        String parentPath = getParentPath(path);
+        int beforeSibling = findNextExistingSiblingStartLine(currentModel, defaultModel, path, parentPath);
+        if (beforeSibling >= 0) {
+            return beforeSibling;
+        }
+
+        if (parentPath == null) {
+            Entry lastRoot = null;
+            for (Entry entry : currentModel.entriesInOrder) {
+                if (getParentPath(entry.path) == null) {
+                    lastRoot = entry;
+                }
+            }
+            return lastRoot == null ? currentLines.size() : lastRoot.endLine + 1;
+        }
+
+        Entry parent = currentModel.entriesByPath.get(parentPath);
+        if (parent == null) {
+            return currentLines.size();
+        }
+        return parent.endLine + 1;
+    }
+
+    private static int findNextExistingSiblingStartLine(
+            DocumentModel currentModel,
+            DocumentModel defaultModel,
+            String path,
+            String parentPath
+    ) {
+        boolean foundSelf = false;
+        for (Entry defaultEntry : defaultModel.entriesInOrder) {
+            if (!isDirectChildOf(defaultEntry.path, parentPath)) {
+                continue;
+            }
+
+            if (!foundSelf) {
+                if (defaultEntry.path.equals(path)) {
+                    foundSelf = true;
+                }
+                continue;
+            }
+
+            Entry existingSibling = currentModel.entriesByPath.get(defaultEntry.path);
+            if (existingSibling != null) {
+                return existingSibling.startLine;
+            }
+        }
+        return -1;
+    }
+
+    private static boolean isDirectChildOf(String path, String parentPath) {
+        String actualParent = getParentPath(path);
+        if (parentPath == null) {
+            return actualParent == null;
+        }
+        return parentPath.equals(actualParent);
+    }
+
+    private static String getParentPath(String path) {
+        int dot = path.lastIndexOf('.');
+        if (dot <= 0) {
+            return null;
+        }
+        return path.substring(0, dot);
+    }
+
+    private static void insertSnippetWithSpacing(List<String> lines, int insertAt, List<String> snippet) {
+        if (snippet.isEmpty()) {
+            return;
+        }
+
+        // Avoid carrying leading/trailing blank lines from the template snippet.
+        int from = 0;
+        int to = snippet.size() - 1;
+        while (from <= to && snippet.get(from).trim().isEmpty()) {
+            from++;
+        }
+        while (to >= from && snippet.get(to).trim().isEmpty()) {
+            to--;
+        }
+        if (from > to) {
+            return;
+        }
+
+        List<String> normalized = new ArrayList<String>(snippet.subList(from, to + 1));
+        int index = insertAt;
+        boolean isTopLevel = isTopLevelSnippet(normalized);
+        if (isTopLevel
+                && index > 0
+                && !lines.get(index - 1).trim().isEmpty()
+                && !normalized.get(0).trim().isEmpty()
+                && (index >= lines.size() || !lines.get(index).trim().isEmpty())) {
+            lines.add(index, "");
+            index++;
+        }
+        lines.addAll(index, normalized);
+    }
+
+    private static boolean isTopLevelSnippet(List<String> snippet) {
+        for (String line : snippet) {
+            KeyLine keyLine = parseKeyLine(line);
+            if (keyLine != null) {
+                return keyLine.indent == 0;
+            }
+        }
+        return false;
+    }
+
+    private static File createBackup(File configFile) throws IOException {
+        String suffix = new SimpleDateFormat("yyyyMMdd-HHmmss", Locale.ROOT).format(new Date());
+        File backup = new File(configFile.getParentFile(), configFile.getName() + ".bak." + suffix);
+        Files.copy(configFile.toPath(), backup.toPath(), StandardCopyOption.REPLACE_EXISTING);
+        return backup;
+    }
+
+    private static DocumentModel parseDocument(List<String> lines) {
+        List<Entry> ordered = new ArrayList<Entry>();
+        Map<String, Entry> byPath = new LinkedHashMap<String, Entry>();
+        ArrayDeque<Entry> stack = new ArrayDeque<Entry>();
+
+        for (int i = 0; i < lines.size(); i++) {
+            KeyLine keyLine = parseKeyLine(lines.get(i));
+            if (keyLine == null) {
+                continue;
+            }
+
+            while (!stack.isEmpty() && keyLine.indent <= stack.peek().indent) {
+                stack.pop();
+            }
+
+            String path = stack.isEmpty() ? keyLine.key : stack.peek().path + "." + keyLine.key;
+            Entry entry = new Entry(path, keyLine.indent, i);
+            ordered.add(entry);
+            byPath.put(path, entry);
+            stack.push(entry);
+        }
+
+        for (Entry entry : ordered) {
+            entry.startLine = findSnippetStartLine(lines, entry.line);
+            entry.endLine = findValueEndLine(lines, entry.line, entry.indent);
+        }
+
+        return new DocumentModel(ordered, byPath);
+    }
+
+    private static int findSnippetStartLine(List<String> lines, int keyLineIndex) {
+        int start = keyLineIndex;
+        for (int i = keyLineIndex - 1; i >= 0; i--) {
+            String trimmed = lines.get(i).trim();
+            if (trimmed.startsWith("#")) {
+                start = i;
+                continue;
+            }
+            break;
+        }
+        return start;
+    }
+
+    private static int findValueEndLine(List<String> lines, int keyLineIndex, int keyIndent) {
+        int end = keyLineIndex;
+        for (int i = keyLineIndex + 1; i < lines.size(); i++) {
+            String line = lines.get(i);
+            KeyLine keyLine = parseKeyLine(line);
+            if (keyLine != null && keyLine.indent <= keyIndent) {
+                break;
+            }
+
+            String trimmed = line.trim();
+            if (!trimmed.isEmpty() && !trimmed.startsWith("#")) {
+                end = i;
+            }
+        }
+        return end;
+    }
+
+    private static KeyLine parseKeyLine(String line) {
+        if (line == null) {
+            return null;
+        }
+
+        String trimmed = line.trim();
+        if (trimmed.isEmpty() || trimmed.startsWith("#") || trimmed.startsWith("-")) {
+            return null;
+        }
+
+        int colon = trimmed.indexOf(':');
+        if (colon <= 0) {
+            return null;
+        }
+
+        String key = trimmed.substring(0, colon).trim();
+        if (key.isEmpty()) {
+            return null;
+        }
+
+        int indent = 0;
+        while (indent < line.length() && line.charAt(indent) == ' ') {
+            indent++;
+        }
+        return new KeyLine(key, indent);
+    }
+
+    public static final class UpdateResult {
+        public final boolean updated;
+        public final int insertedPaths;
+        public final String backupFileName;
+
+        private UpdateResult(boolean updated, int insertedPaths, String backupFileName) {
+            this.updated = updated;
+            this.insertedPaths = insertedPaths;
+            this.backupFileName = backupFileName;
+        }
+
+        private static UpdateResult noChange() {
+            return new UpdateResult(false, 0, "");
+        }
+
+        private static UpdateResult updated(int insertedPaths, String backupFileName) {
+            return new UpdateResult(true, insertedPaths, backupFileName);
+        }
+    }
+
+    private static final class DocumentModel {
+        private final List<Entry> entriesInOrder;
+        private final Map<String, Entry> entriesByPath;
+
+        private DocumentModel(List<Entry> entriesInOrder, Map<String, Entry> entriesByPath) {
+            this.entriesInOrder = entriesInOrder;
+            this.entriesByPath = entriesByPath;
+        }
+    }
+
+    private static final class Entry {
+        private final String path;
+        private final int indent;
+        private final int line;
+        private int startLine;
+        private int endLine;
+
+        private Entry(String path, int indent, int line) {
+            this.path = path;
+            this.indent = indent;
+            this.line = line;
+            this.startLine = line;
+            this.endLine = line;
+        }
+    }
+
+    private static final class KeyLine {
+        private final String key;
+        private final int indent;
+
+        private KeyLine(String key, int indent) {
+            this.key = key;
+            this.indent = indent;
+        }
+    }
+}

src/main/resources/config.yml

@@ -1,30 +1,266 @@
-# OpenAI API key setting
+# ======================================================
+# Language Settings
+# 语言设置
+# ======================================================
+# Available languages: en, zh
+# 可用语言：en（英文）, zh（中文）
+# This only affects plugin prompts, not the model's answer language.
+# 该项仅影响插件提示语言，不影响模型回答语言。
+language: "en"
+
+# ======================================================
+# API Configuration
+# API 设置
+# ======================================================
 api:
-  # Your OpenAI API key, used for authentication
+  # Your OpenAI API keys, used for authentication
   # To obtain an API key, visit https://platform.openai.com/account/api-keys and create a new API key
-  key: "sk-your_openai_api_key"
+  # 你的 OpenAI API key，用于身份验证
+  # 获取 API key 的方法：访问 https://platform.openai.com/account/api-keys 并创建一个新的 API key
+  # Placeholder keys such as sk-your_openai_api_key_1 are treated as invalid.
+  # 占位 key（如 sk-your_openai_api_key_1）会被视为无效。
+  keys:
+    - "sk-your_openai_api_key_1"
+    # You can add multiple API keys below
+    # 可以添加多个API key
+    # - "sk-your_openai_api_key_2"
+    # - "sk-your_openai_api_key_3"
+
+  # API key selection method: "round_robin" or "random"
+  # Round Robin: Use each API key in turn
+  # Random: Randomly select an API key
+  # API key 选择方法："round_robin"（轮询）或 "random"（随机）
+  # 轮询：依次使用每个API key
+  # 随机：随机选择一个API key
+  # Use round_robin for steadier distribution in production.
+  # 生产环境更推荐 round_robin，分配更均匀。
+  selection_method: "round_robin"
+
   # The base URL for the OpenAI API, used to construct requests
+  # If you cannot access the official API, you can use a proxy service
+  # OpenAI API 的基础 URL，用于构建请求
+  # 如果你无法访问官方API，可以使用代理服务
+  # Keep the /v1 path in the URL.
+  # URL 保留 /v1 路径。
   base_url: "https://api.openai.com/v1"
+
+  # HTTP timeouts (milliseconds)
+  # HTTP 超时（毫秒）
+  # connect_timeout_ms controls connection phase, timeout_ms controls total request time.
+  # connect_timeout_ms 控制建连阶段，timeout_ms 控制整次请求耗时。
+  # Increase carefully on unstable networks to avoid slow failure feedback.
+  # 网络不稳定时可适当调大，但过大可能导致失败反馈变慢。
+  connect_timeout_ms: 10000
+  timeout_ms: 30000
+
+  # Thread pool size for API requests
+  # API 请求线程池大小
+  # Higher value increases concurrency but also upstream pressure.
+  # 值越大并发越高，但也会增加上游 API 压力。
+  # Tune together with queue.max_in_flight.
+  # 建议与 queue.max_in_flight 配合调优。
+  thread_pool_size: 4
+
+# ======================================================
+# Model Configuration
+# 模型设置
+# ======================================================
 # List of supported models
+# 支持的模型列表
+# This is a selectable whitelist; actual availability depends on your provider.
+# 这里只是可选白名单，具体是否可用取决于服务商支持情况。
 models:
   # OpenAI ChatGPT
-  - "gpt-3.5-turbo"
   - "gpt-4"
+  - "gpt-4-turbo"
+  - "gpt-4-turbo-preview"
+  - "gpt-4o"
+  - "gpt-4o-mini"
+  # Anthropic Claude
+  # - "claude-3-opus"
+  # - "claude-3-5-sonnet"
+  # Deepseek
+  # - "deepseek-coder"
+  # - "deepseek-reasoner"
+  # And more...
+  # 以及更多...
+
 # The default model to use
-default_model: "gpt-3.5-turbo"
+# 默认使用的模型
-# Message settings
+# Use a cost-effective model as default for daily usage.
-messages:
+# 日常建议选择更稳定且成本更低的模型作为默认模型。
-  reload: "Configuration reloaded successfully!"
+default_model: "gpt-4o-mini"
-  help: "===== MineChatGPT Help ====="
+
-  help_ask: "/chatgpt <text> - Ask ChatGPT a question."
+# ======================================================
-  help_reload: "/chatgpt reload - Reload the configuration file."
+# Conversation Settings
-  help_model: "/chatgpt model <model_name> - Switch to a different model."
+# 对话设置
-  help_modellist: "/chatgpt modellist - List available models."
+# ======================================================
-  usage: "Usage: /chatgpt model <model_name>"
+conversation:
-  model_switch: "Model switched to %s"
+  # Continuous conversation switch
-  chatgpt_error: "Failed to contact ChatGPT."
+  # When enabled, the plugin will remember the conversation history
-  chatgpt_response: "ChatGPT: %s"
+  # 连续对话开关
-  question: "Question: %s"
+  # 启用时，插件将记住对话历史
-  invalid_model: "Invalid model. Use /chatgpt modellist to see available models."
+  # Enabling improves continuity but increases token usage.
-  available_models: "Available models:"
+  # 开启后连贯性更好，但 token 消耗会增加。
-  no_permission: "You do not have permission to use this command. Required permission: %s"
+  context_enabled: false
+
+  # Maximum number of historical records retained
+  # Increasing this value will consume more memory
+  # 最大历史记录保留数量
+  # 增加此值将消耗更多内存
+  # Usually 8~20 is a practical range.
+  # 通常 8~20 是较实用的范围。
+  max_history_size: 10
+
+  # Token budget for context (approximate)
+  # 上下文 token 预算（近似）
+  # Exceeding this budget may trigger trimming or summarization.
+  # 超过预算会触发上下文裁剪或总结。
+  max_context_tokens: 2000
+
+  # Reserve tokens for the model output (approximate)
+  # 为模型输出预留 token（近似）
+  # Too low may truncate replies; too high reduces usable context.
+  # 值太小可能截断回复，值太大则压缩可用上下文。
+  reserve_completion_tokens: 400
+
+  # Automatic summarization to keep context within budget
+  # 自动总结以控制上下文长度
+  summarization:
+    # If disabled, long sessions are more likely to hit context limits.
+    # 关闭后长会话更容易触发上下文超限。
+    enabled: true
+    # Model used for summarization
+    # 用于总结的模型
+    # Prefer a faster/cheaper model for summarization.
+    # 建议用更快/更便宜的模型做总结。
+    model: "gpt-4o-mini"
+    # When estimated context tokens exceed this value, summarization may run
+    # 当估算的上下文 tokens 超过该值时可能触发总结
+    # Usually set slightly lower than max_context_tokens (e.g. 100~300 lower).
+    # 通常设为略低于 max_context_tokens（如低 100~300）。
+    trigger_tokens: 1800
+    # Keep the last N messages (user/assistant) after summarization
+    # 总结后保留最近 N 条消息（user/assistant）
+    # Higher value keeps more detail but uses more tokens.
+    # 值越大保留细节越多，但占用也更高。
+    keep_last_messages: 6
+
+# ======================================================
+# Rate Limit / Queue Settings
+# 限流 / 排队 设置
+# ======================================================
+rate_limit:
+  # Master switch for cooldown/token-bucket limiting.
+  # 冷却与令牌桶限流总开关。
+  enabled: true
+  # Mode: "requests" | "tokens" | "both"
+  # 模式："requests" | "tokens" | "both"
+  # requests = limit by request count
+  # tokens = limit by token budget
+  # both = enforce both limits (recommended)
+  # requests = 仅按请求次数
+  # tokens = 仅按 token 预算
+  # both = 两者同时生效（推荐）
+  mode: "both"
+
+  # Cooldown per user after accepting a request (milliseconds)
+  # 每个用户请求冷却（毫秒）
+  # Helps reduce short burst command spam from one player.
+  # 用于抑制单玩家短时间刷命令。
+  cooldown_ms: 1000
+
+  # Token estimation method: "approx_chars"
+  # token 估算方法："approx_chars"
+  # This estimator is approximate by design.
+  # 该估算器是近似估算。
+  token_estimator: "approx_chars"
+
+  # If token-mode is enabled, assume this many completion tokens per request for limiting
+  # 如果启用 token 限流，按该值假设每次回复的 completion tokens（用于限流）
+  # Increase this if users often ask for long outputs.
+  # 用户经常要求长回复时可适当提高该值。
+  assumed_completion_tokens: 300
+
+  per_user:
+    # Per-user request refill rate (requests/min).
+    # 单用户请求额度恢复速率（每分钟）。
+    requests_per_minute: 6
+    # Per-user request burst capacity.
+    # 单用户请求突发容量。
+    burst_requests: 3
+    # Per-user token refill rate (tokens/min).
+    # 单用户 token 额度恢复速率（每分钟）。
+    tokens_per_minute: 4000
+    # Per-user token burst capacity.
+    # 单用户 token 突发容量。
+    burst_tokens: 2000
+
+  global:
+    # Global request refill rate (requests/min).
+    # 全局请求额度恢复速率（每分钟）。
+    requests_per_minute: 60
+    # Global request burst capacity.
+    # 全局请求突发容量。
+    burst_requests: 20
+    # Global token refill rate (tokens/min).
+    # 全局 token 额度恢复速率（每分钟）。
+    tokens_per_minute: 100000
+    # Global token burst capacity.
+    # 全局 token 突发容量。
+    burst_tokens: 20000
+
+queue:
+  # If enabled, requests can wait in queue instead of immediate rejection.
+  # 开启后，请求可进入队列等待，而不是直接拒绝。
+  enabled: true
+  # Maximum queued jobs globally
+  # 全局队列最大长度
+  # New jobs are rejected after this limit.
+  # 达到上限后新请求会被拒绝。
+  max_size: 100
+  # Maximum queued jobs per user
+  # 每个用户队列最大长度
+  # Prevent one player from occupying all queue slots.
+  # 防止单玩家占满队列。
+  max_per_user: 3
+  # Maximum in-flight (concurrent) API requests
+  # 同时进行中的 API 请求上限
+  # Usually set <= api.thread_pool_size.
+  # 通常建议不高于 api.thread_pool_size。
+  max_in_flight: 4
+  # How many jobs to dispatch per tick (Spigot main thread)
+  # 每 tick 尝试派发多少个任务（主线程）
+  # Higher value drains queue faster but increases main-thread scheduling pressure.
+  # 值越大清队越快，但主线程调度压力更高。
+  dispatch_per_tick: 2
+
+# ======================================================
+# Character Settings
+# 角色设置
+# ======================================================
+characters:
+  # Format:
+  # Character Name: "Character Prompt"
+  # The prompt will be sent to the AI as a system message
+  # 格式：
+  # 角色名称: "角色提示词"
+  # 提示词将作为系统消息发送给AI
+  # Keep prompts clear and concise to avoid unnecessary token overhead.
+  # 建议提示词简洁明确，避免无谓 token 开销。
+
+  # Default character
+  ChatGPT: "You are a helpful assistant."
+
+  # You can add more characters below
+  # 你可以在下方添加更多角色
+  # Example:
+  # Minecraft: "You are a Minecraft expert who helps players with game mechanics and building ideas."
+
+# ======================================================
+# Other Settings
+# 其他设置
+# ======================================================
+# If you don't know what this is, don't change it
+# 如果你不知道这是什么，请不要动
+debug: false

src/main/resources/config_zh.yml

@@ -1,30 +0,0 @@
-# API 相关设置
-api:
-  # 你的 OpenAI API key，用于身份验证
-  # 获取 API key 的方法：访问 //platform.openai.com/account/api-keys 并创建一个新的 API key
-  key: "your_openai_api_key"
-  # OpenAI API 的基础 URL，用于构建请求
-  base_url: "https://api.openai.com/v1"
-# 支持的模型列表
-models:
-  # OpenAI ChatGPT
-  - "gpt-3.5-turbo"
-  - "gpt-4"
-# 默认使用的模型
-default_model: "gpt-3.5-turbo"
-# 消息相关设置
-messages:
-  reload: "已重新加载配置文件！"
-  help: "===== MineChatGPT 帮助 ====="
-  help_ask: "/chatgpt <text> - 向ChatGPT提问"
-  help_reload: "/chatgpt reload - 重新加载配置文件"
-  help_model: "/chatgpt model <model_name> - 切换至其他模型"
-  help_modellist: "/chatgpt modellist - 可用的模型列表"
-  usage: "输入： /chatgpt model <model_name>"
-  model_switch: "已切换至模型 %s"
-  chatgpt_error: "无法联系ChatGPT。"
-  chatgpt_response: "ChatGPT: %s"
-  question: "你: %s"
-  invalid_model: "模型无效。使用 /chatgpt modellist 查看可用模型。"
-  available_models: "可用模型列表："
-  no_permission: "你没有权限使用这个指令。需要的权限：%s"

src/main/resources/lang/en.yml

@@ -0,0 +1,38 @@
+# English language file for MineChatGPT
+
+messages:
+  reload: "&aConfiguration reloaded successfully!"
+  clear: "&aConversation history has been cleared!"
+  help: "&e===== MineChatGPT Help ====="
+  help_ask: "&e/chatgpt <text> - Ask ChatGPT a question."
+  help_reload: "&e/chatgpt reload - Reload the configuration file."
+  help_model: "&e/chatgpt model <model_name> - Switch to a different model."
+  help_modellist: "&e/chatgpt modellist - List available models."
+  help_context: "&e/chatgpt context - Toggle context mode."
+  help_clear: "&e/chatgpt clear - Clear conversation history."
+  help_character: "&e/chatgpt character [character_name] - List or switch to a character."
+  help_stats: "&e/chatgpt stats - Show token usage statistics."
+  context_toggle: "&eContext is now %s."
+  context_toggle_enabled: "&aenabled"
+  context_toggle_disabled: "&cdisabled"
+  current_model_info: "&eCurrent model: %s. Use /chatgpt model <model_name> to switch models."
+  model_switch: "&aModel switched to %s"
+  chatgpt_error: "&cFailed to contact ChatGPT."
+  no_api_key: "&cNo API key configured. Please set api.keys in config.yml."
+  chatgpt_response: "&b%s: %s"
+  question: "&bYou: %s"
+  queued: "&eYour request has been queued. Position: %s"
+  queue_full: "&cQueue is full. Please try again later."
+  queue_full_user: "&cYou have too many pending requests. Please wait."
+  cooldown: "&cYou're sending requests too fast. Please wait %s ms."
+  rate_limited: "&cRate limited. Please try again later."
+  stats_header: "&e===== MineChatGPT Stats ====="
+  stats_global: "&eGlobal tokens: %s (prompt=%s, completion=%s), requests=%s"
+  stats_user: "&eYour tokens: %s (prompt=%s, completion=%s), requests=%s"
+  stats_reset: "&aStats reset."
+  character_switched: "&aSwitched to character: %s"
+  available_characters: "&eAvailable characters:"
+  invalid_character: "&cInvalid character. Use /chatgpt character to list available characters."
+  invalid_model: "&cInvalid model. Use /chatgpt modellist to see available models."
+  available_models: "&eAvailable models:"
+  no_permission: "&cYou do not have permission to use this command. Required permission: %s"

src/main/resources/lang/zh.yml

@@ -0,0 +1,38 @@
+# 中文语言文件 - MineChatGPT
+
+messages:
+  reload: "&a已重新加载配置文件！"
+  clear: "&a对话历史已清空！"
+  help: "&e===== MineChatGPT 帮助 ====="
+  help_ask: "&e/chatgpt <text> - 向ChatGPT提问"
+  help_reload: "&e/chatgpt reload - 重新加载配置文件"
+  help_model: "&e/chatgpt model <model_name> - 切换至其他模型"
+  help_modellist: "&e/chatgpt modellist - 可用的模型列表"
+  help_context: "&e/chatgpt context - 切换连续对话模式"
+  help_clear: "&e/chatgpt clear - 清空对话历史"
+  help_character: "&e/chatgpt character [character_name] - 列出或切换角色"
+  help_stats: "&e/chatgpt stats - 查看 token 消耗统计"
+  context_toggle: "&e连续对话模式已%s。"
+  context_toggle_enabled: "&a开启"
+  context_toggle_disabled: "&c关闭"
+  current_model_info: "&e当前模型：%s，输入 /chatgpt model <model_name> 来切换模型。"
+  model_switch: "&a已切换至模型 %s"
+  chatgpt_error: "&c无法联系ChatGPT。"
+  no_api_key: "&c未配置 API Key，请在 config.yml 的 api.keys 中设置。"
+  chatgpt_response: "&b%s: %s"
+  question: "&b你: %s"
+  queued: "&e你的请求已进入队列，当前位置：%s"
+  queue_full: "&c队列已满，请稍后再试。"
+  queue_full_user: "&c你有太多等待中的请求，请先等待。"
+  cooldown: "&c请求过快，请等待 %s 毫秒。"
+  rate_limited: "&c触发限流，请稍后再试。"
+  stats_header: "&e===== MineChatGPT 统计 ====="
+  stats_global: "&e全局 tokens：%s（prompt=%s，completion=%s），请求数=%s"
+  stats_user: "&e你的 tokens：%s（prompt=%s，completion=%s），请求数=%s"
+  stats_reset: "&a统计已重置。"
+  character_switched: "&a已切换至角色： %s"
+  available_characters: "&e可用的角色列表："
+  invalid_character: "&c无效的角色。使用 /chatgpt character 查看所有可用的角色。"
+  invalid_model: "&c模型无效。使用 /chatgpt modellist 查看可用模型。"
+  available_models: "&e可用模型列表："
+  no_permission: "&c你没有权限使用这个指令。需要的权限：%s"

src/main/resources/plugin.yml

@@ -2,12 +2,13 @@ name: MineChatGPT
 version: '${version}'
 main: com.ddaodan.MineChatGPT.Main
 author: ddaodan
+folia-supported: true
 description: A Spigot plugin for interacting with ChatGPT
 
 commands:
   chatgpt:
     description: Interact with ChatGPT
-    usage: /chatgpt <question>
+    usage: /chatgpt <text>
 permissions:
   minechatgpt.use:
     description: Allows using the /chatgpt command to interact with AI
@@ -21,3 +22,15 @@ permissions:
   minechatgpt.modellist:
     description: Allows listing available models
     default: op
+  minechatgpt.context:
+    description: Allows toggling context mode
+    default: true
+  minechatgpt.clear:
+    description: Allows clearing conversation history
+    default: true
+  minechatgpt.character:
+    description: Allows switching characters
+    default: true
+  minechatgpt.stats:
+    description: Allows viewing/resetting usage statistics
+    default: op