界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
效率与协作 · SKILL.md

persistence-rag-workspaces

persistence-rag-workspaces

【效率与协作】Maintain LangBot persistence, Alembic migrations, Workspace tenancy, cloud directory state, RAG knowledge bases, vector backends, storage, monitoring, and telemetry workflows.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: persistence-rag-workspaces description: "Maintain LangBot persistence, Alembic migrations, Workspace tenancy, cloud directory state, RAG knowledge bases, vector backends, storage, monitoring, and telemetry workflows." disable-model-invocation: true metadata: disco-role: operating license: Apache 2.0

Persistence, RAG, Workspaces, and Monitoring

Use this sub-skill for database models, migrations, tenant scoping, Workspace collaboration, cloud directory/launch behavior, RAG/knowledge-base operations, vector backends, file storage, monitoring exports, and telemetry.

Read First

Hard Rules

  • New schema changes use Alembic under the current Alembic versions directory; do not add new legacy dbmXXX migrations.
  • Tenant services must accept a RequestContext or trusted execution context and fail closed when context is missing.
  • API keys are bound to one Workspace; do not let request headers switch an API key's tenant.
  • Vector/database service-backed tests are optional unless the task changes that backend; SQLite and unit tests cover the general path.

Focused Checks

python scripts/select_langbot_checks.py persistence-rag
uv run pytest tests/integration/persistence/test_migrations.py -q --tb=short
uv run pytest tests/unit_tests/vector/test_mgr.py tests/unit_tests/vector/test_vdb_filter_conversion.py -q --tb=short
uv run pytest tests/unit_tests/storage/test_storage_manager.py tests/unit_tests/rag/test_runtime_service.py -q --tb=short
我的收藏