Commit Graph

64 Commits

Author SHA1 Message Date
jxxghp
ef0f4f7ef4 feat(agent): improve Windows shell compatibility 2026-09-01 17:51:15 +08:00
jxxghp
41c54d138d refactor(agent): relocate static resources 2026-09-01 11:55:21 +08:00
jxxghp
0c05956814 refactor(architecture): enforce module naming policy 2026-08-29 22:17:17 +08:00
jxxghp
dbdf7d0582 refactor(agent): narrow package compatibility surfaces 2026-08-29 14:29:42 +08:00
jxxghp
c000c4fff3 refactor(agent): split llm provider owners 2026-08-29 13:57:48 +08:00
jxxghp
92e745fc97 refactor(architecture): split chain owners and close runtime boundaries 2026-08-28 22:26:19 +08:00
jxxghp
9dbe424c3d refactor(config): retire RuntimeSettingsCompat host usage 2026-08-26 15:55:21 +08:00
jxxghp
7c9d53c6f6 refactor: unify llm provider runtime access 2026-08-24 07:15:02 +08:00
jxxghp
cb3c5f008f refactor: unify service configuration boundary 2026-08-24 05:12:21 +08:00
jxxghp
8555d0da3e refactor: use explicit system config getter 2026-08-24 04:04:35 +08:00
jxxghp
41ee5aacdc refactor: route agent llm settings through runtime 2026-08-23 02:49:00 +08:00
InfinityPacer
23f5d59c74 feat: 使用 uv 锁定主程序依赖并强化插件恢复边界 (#6364) 2026-08-20 12:17:19 +08:00
jxxghp
8472bcff43 refactor: 收口 V3 分层架构与插件兼容边界 2026-08-18 13:22:02 +08:00
jxxghp
cca99bd421 ci: fetch models.dev catalog during packaging 2026-08-18 13:18:36 +08:00
jxxghp
54f093a099 Merge remote-tracking branch 'origin/v3' into v3
# Conflicts:
#	app/api/endpoints/agent.py
#	app/api/endpoints/anthropic.py
#	app/api/endpoints/openai.py
#	app/chain/__init__.py
#	app/chain/message.py
#	app/chain/site.py
#	app/chain/subscribe.py
#	app/chain/transfer.py
#	app/modules/discord/__init__.py
#	app/modules/qqbot/__init__.py
#	app/modules/slack/__init__.py
#	app/modules/telegram/__init__.py
#	app/modules/wechat/__init__.py
#	app/runtime/extensions/module_manager.py
#	app/runtime/extensions/service_registry.py
#	tests/test_agent_interaction.py
#	tests/test_slash_command_interactions.py
#	tests/test_web_agent_stream.py
2026-08-16 19:44:43 +08:00
jxxghp
240a4dffe6 refactor(schemas): 统一 message/notification 命名边界,旧名收敛至兼容映射表
- notification 域:渠道能力(MessageChannel→NotificationChannel、ChannelCapability* 迁入 notification.py)
- message 域:消息收发(Notification→Message、NotificationType→MessageType、CommingMessage→IncomingMessage、NotificationHistoryItem→MessageHistoryItem、NotificationClear*→MessageClear*)
- Agent 工具契约:send_notification_message→send_message、notification_callback→message_callback
- 源码不保留旧名物理别名,旧导入经 app/runtime/compat/manifest.py SYMBOL_ALIASES 惰性解析
- API 路径与持久化键冻结不变,前端零改动
- 新增兼容守护测试与 docs/rules/07 命名边界规范
2026-08-16 19:32:20 +08:00
InfinityPacer
5b367011c5 refactor(agent): 按需加载 Agent 运行时 (#6336) 2026-08-16 19:25:24 +08:00
jxxghp
441ec9475e refactor(api): LLM 提供商管理端点收敛为通用 manage 接口,端点层零特色
- schemas 新增 LlmProviderAction 公共动作词汇表
- LLMProviderManager 新增 provider_manage 统一入口,默认值填充、
  API Key 豁免判断、密钥脱敏与错误归因改写全部下沉封闭,
  不再硬编码 chatgpt/github-copilot 等产品名
- endpoint 收敛为 POST /llm/manage(ManageRequest 透传);
  OAuth 回跳地址由端点按具名回调路由统一构造后注入动作参数,
  /provider-auth/callback/{provider_id} 因浏览器回跳协议约束保留具名路由
- 新增 14 项 provider_manage 契约守护测试,
  替换原针对端点函数的两个旧测试文件
2026-08-16 07:50:16 +08:00
Aqr-K
8a11214a43 refactor(db): 修复异步连接池无界增长,并完成 SQLAlchemy 2.0 迁移与分层归位 (#6320) 2026-08-15 06:58:38 +08:00
jxxghp
369d7d6448 refactor: reorganize backend module boundaries 2026-08-14 19:53:33 +08:00
jxxghp
7b3444c366 refactor backend module architecture 2026-08-14 15:45:38 +08:00
InfinityPacer
ce3508730b fix(agent): 规范化模型窗口 Profile (#6289) 2026-08-13 13:51:22 +08:00
jxxghp
e6d7d57e6f fix(agent): 修正 OpenAI 兼容 API Key 提示 (#6243) 2026-08-09 16:15:14 +08:00
jxxghp
a52e1fdc1c feat(agent): improve prompt cache hit rate 2026-08-06 23:34:29 +08:00
jxxghp
a23ac6c56d feat: 支持模型服务端联网搜索 2026-08-05 19:19:26 +08:00
jxxghp
33a97eb2c8 feat: 增加 MoviePilot 选项 (#6212) 2026-07-30 13:33:08 +08:00
freeman
572299a45e feat(llm): 新增 Amazon Bedrock 提供商,支持 AK/SK 与 Bedrock API Key 双认证 (#6130) 2026-07-16 19:27:06 +08:00
jxxghp
a71d3ea03f feat: support llm temperature override 2026-07-05 13:04:51 +08:00
jxxghp
0cd049bfc2 Refactor movie pilot config and test coverage 2026-06-23 10:05:45 +08:00
jxxghp
3306d196b7 Refine existing implementation 2026-06-22 18:21:20 +08:00
jxxghp
90efb204a1 refactor: expose text content extraction helper 2026-06-21 12:56:23 +08:00
jxxghp
68f18db374 Handle invalid UTF-8 with replacement decoding 2026-06-21 09:56:34 +08:00
jxxghp
ac5374c244 feat: enhance audio capability logging for transcription and synthesis 2026-06-01 12:55:28 +08:00
jxxghp
13b2163788 chore: add noqa for specific lines, update docstring, improve logging and variable naming 2026-05-31 17:49:05 +08:00
jxxghp
ac09ce5230 fix: handle empty ChatGPT responses output 2026-05-31 08:14:31 +08:00
jxxghp
492e3c333b fix: unify User-Agent header usage with settings.USER_AGENT 2026-05-30 09:49:29 +08:00
jxxghp
1f7fb304dd feat: add LLM proxy toggle 2026-05-27 06:57:09 +08:00
jxxghp
896631d63e fix(agent): enable voice replies for supported channels 2026-05-26 20:14:56 +08:00
jxxghp
e00e18f31e fix: transcode unsupported stt audio formats 2026-05-26 10:00:32 +08:00
jxxghp
c7965edd47 fix: use responses api for chatgpt reasoning models 2026-05-26 09:31:00 +08:00
jxxghp
821bd3decd fix: use explicit LLM user agent config 2026-05-26 08:24:58 +08:00
jxxghp
b65c8dcfe0 feat: support llm user agent 2026-05-26 08:20:02 +08:00
jxxghp
0e8bcb4df6 fix: patch gemini thought_signature enforcement to cover all function calls
The upstream _parse_chat_history enforcement code uses a first_fc_seen
flag that only adds DUMMY_THOUGHT_SIGNATURE to the first function_call
without thought_signature. Parallel function calls (position 2+) remain
unpatched, causing Gemini API 400 errors for all Gemini 2.5+ models.

Additionally, _is_gemini_3_or_later only matches 'gemini-3', missing
Gemini 2.5 models entirely.

This patch:
1. Extends _is_gemini_3_or_later to also match gemini-2.5 models
2. Wraps _parse_chat_history to ensure ALL function_call parts in ALL
   model messages have thought_signature (not just the first one)
2026-05-25 13:53:02 +08:00
jxxghp
766d2699ea feat: support MiniMax audio provider 2026-05-25 11:42:57 +08:00
jxxghp
79539760da fix: bound long-lived cache state 2026-05-24 18:03:42 +08:00
jxxghp
dc73d61682 feat: add china operator llm providers 2026-05-24 09:30:41 +08:00
jxxghp
7b27b7fd16 feat: add extensible agent audio capabilities 2026-05-14 19:37:13 +08:00
jxxghp
4322f8a3c1 fix: preserve reasoning content for compatible llms 2026-05-14 14:01:53 +08:00
jxxghp
fc2c77fbf1 fix(agent): refresh LLM runtime config on each call
Read the latest LLM connection settings when building runtime clients so Web updates take effect immediately instead of reusing module-import defaults.

Closes #5757
2026-05-12 18:48:31 +08:00
jxxghp
e4c5a4f232 feat(provider): add kuaishou-wanqing endpoint with base URL presets and manual model input 2026-05-08 15:16:29 +08:00