DDSRem
e1b703aacc
refactor(agent): centralize graph resource ownership ( #7062 )
2026-10-18 21:49:57 +08:00
DDSRem
2de976f6b3
Merge pull request #7050 from DDSRem-Dev/codex-comment-migration
...
docs: 统一全项目中文注释和结构化契约说明
2026-10-17 19:26:29 +08:00
jxxghp
6c9ebd01f3
fix(agent): close owned LLM clients with their lifecycle
...
Refs #6901
2026-10-07 09:18:40 +08:00
jxxghp
f803e6cc5c
feat(agent): add persistent Python read-only orchestration
2026-09-30 20:16:53 +08:00
jxxghp
fcd4cd1725
feat(agent): align execution guardrails with Hermes
2026-09-30 18:51:03 +08:00
jxxghp
3eb339b8e0
feat(agent): unify memory retrieval and activity history
2026-09-16 18:24:30 +08:00
jxxghp
6fe5bb42c2
fix(agent): preserve queued messages and long context
2026-09-11 14:27:27 +08:00
jxxghp
21e154d6d2
feat(agent): deliver browser screenshots as model-visible images
2026-09-10 02:17:01 +08:00
jxxghp
5fa5b414aa
feat(agent): harden tool outcomes, recovery and discovery
2026-09-09 19:55:21 +08:00
jxxghp
c0ed78e168
feat(agent): strengthen planning, tool discovery and recovery
2026-09-09 19:37:11 +08:00
jxxghp
49414ffad2
test: 清理重复单元测试并记录审计依据
2026-09-06 07:58:07 +08:00
jxxghp
ff01449cdd
fix(api): humanize user-facing error messages
2026-09-04 17:55:56 +08:00
jxxghp
d512f528be
refactor(api): make collection pagination explicit
2026-09-01 09:13:56 +08:00
jxxghp
fdba2c7d64
refactor(agent): unify tools behind API gateway and provider skills
2026-08-31 16:57:54 +08:00
jxxghp
0c05956814
refactor(architecture): enforce module naming policy
2026-08-29 22:17:17 +08:00
jxxghp
3a33da0b38
refactor(agent): split manager runtime owners
2026-08-29 14:26:12 +08:00
jxxghp
7b3444c366
refactor backend module architecture
2026-08-14 15:45:38 +08:00
InfinityPacer
8dc6774d47
fix(agent): preserve context during request compaction ( #6300 )
2026-08-13 17:58:31 +08:00
InfinityPacer
228de7bd67
fix(agent): compact oversized final model requests ( #6299 )
2026-08-13 17:26:13 +08:00
InfinityPacer
527ceff5d1
fix(agent): preserve context when summarization fails ( #6286 )
2026-08-13 10:07:56 +08:00
jxxghp
78ddd6093f
Improve streaming tool summary counts
2026-06-21 18:49:27 +08:00
InfinityPacer
437baec620
test: 测试套件自隔离与全量离线化（collection 清零 + 杜绝真实网络） ( #5873 )
2026-06-02 12:23:08 +08:00
jxxghp
c89df496a5
feat(agent): add ToolTag-based tags to all agent tools; implement tags.py for unified tool capability tagging
2026-05-31 18:30:39 +08:00
jxxghp
53bf81aede
refactor: rename MoviePilotToolSelectorMiddleware to ToolSelectorMiddleware and enhance tool selection logic
2026-04-30 19:05:49 +08:00
jxxghp
afcc071d07
feat: optimize tool selection middleware to cache and reuse tool selection per agent run
...
- Refactor MoviePilotToolSelectorMiddleware to perform tool selection once per agent execution and cache the result in state, avoiding redundant LLM calls for each model round.
- Add abefore_agent to select tools at the start of agent execution and store selected tool names in state.
- Update awrap_model_call to reuse cached tool selection from state for subsequent model calls.
- Enhance test coverage for tool selection caching and reuse logic.
- Improve error logging in skill version extraction.
2026-04-30 18:29:54 +08:00
jxxghp
53f6897d62
feat: ensure essential tools are always included in LLM tool selection and update tests
...
- Add mechanism to always include core tools (e.g., file operations, command execution) in LLMToolSelectorMiddleware
- Update MoviePilotToolFactory to provide filtered always-include tool names based on loaded tools
- Set default LLM_MAX_TOOLS to 30 in config
- Refactor agent initialization to support always_include parameter
- Enhance tests to cover always_include logic and async agent creation
2026-04-30 13:04:52 +08:00
jxxghp
344280cd61
Refactor agent persona runtime layering
2026-04-29 14:12:47 +08:00
jxxghp
221eb21694
refine internal middleware llm usage for streaming agents
...
Use a non-streaming model for middleware-only calls so internal outputs do not leak into user streams and model-based middleware stays consistent.
2026-04-27 06:55:41 +08:00
