严格检查全部宿主源码，清零类型错误预算并移除按模块类型豁免。
按原因、所属方法、错误码及语法指纹锁定必要兼容例外和站点类型桩，
CI 阻断新增忽略、检查范围退化、本地桩替代真实接口及新的类型错误。
修复空值、异步和供应方类型边界，保持插件与 API 兼容合同，
并补充存储、订阅、通知、Agent 和工作流边界的回归覆盖。
Keep only install, usage, development-workflow and architecture-boundary
documents; remove the one-off refactor plans, audits and ledgers that had
already served their purpose.
- delete docs/refactor plus the agent API surface audit generator and its
drift-only tests; the fixed route registry remains the source of truth
- delete internal implementation notes (agent, music, notification, transfer,
subscription activity, docker startup, V3t runtime governance)
- delete site adapter capture/release shims and the ADR records; external
links point at MoviePilot-Site-Collector instead
- keep docs/architecture-overview.md and gate its dependency and event
figures against the generated fixtures
- drop doc-only assertions from the architecture baseline test and retarget
remaining references at code and rule documents
Use the music media type as the downloader category only when music has no library classification. Preserve configured categories and qBittorrent automatic-management settings, with SDK-boundary regression coverage.
Refs #7028
The recognize endpoint returned an empty context when no media matched,
so the recognition shortcut could not show the applied custom words or
the processed title, which made failures hard to diagnose. Always
return the parsed meta_info (including apply_words) and leave media_info
empty when the lookup fails.
Update the REST API docs and the generated agent skill note accordingly.
Refs #6997
* feat(agent): scripted generation of agent skill docs with operation/field notes
Generate skills/moviepilot-api API docs from scripts/generate_agent_skill_docs.py
instead of hand-editing:
- surface canonical MediaType agent keys in the API contract
- add subscribeactivity table note and global operation notes
- preserve upstream search.torrents execution semantics via field notes
Co-Authored-By: pavel2 <pavelyuan468@gmail.com>
* fix(agent): preserve generated skill contracts and English guidance
---------
Co-authored-by: jxxghp <jxxghp@gmail.com>
Run `ruff check app tests scripts --fix` with the configured rule set
(E4/E7/E9/F/I): import sorting (I001), f-strings without placeholders
(F541), lambda/variable/unused-import cleanups that ruff marks safe.
Nine files whose only ruff debt sits next to pre-existing pylint
import-resolution errors (lazy exports, sys.path script imports) or
whose "unused" imports are reached through attribute access by tests are
left untouched so the changed-file pylint gate and the benchmark tests
stay green; their diagnostics remain in the baseline.
app/startup/lifecycle/__init__.py and app/main.py import business
modules after cache composition, import-path and stdio setup on purpose
(the source comments document the contract), so declare E402
per-file-ignores for them instead of adding per-line noqa.
Ruff ratchet watermark drops from 498 to 100; sync the baseline fixture
and the optimization checklist figure.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
