From Retrieval Core
Use when deciding HOW to find information — choosing or composing search modalities (lexical, structural, structured-data, history, semantic/RAG, graph) for a query and corpus.
How this skill is triggered — by the user, by Claude, or both
Slash command
/retrieval-core:retrieval-strategyThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Pick a modality by **what you know about the query × what you know about the corpus.**
Pick a modality by what you know about the query × what you know about the corpus.
This skill is portable across GitHub Copilot, APM, and Claude Code. Each host
installs it from the mbeacom/context-kit marketplace — register the marketplace
(copilot plugin marketplace add, apm marketplace add, or
/plugin marketplace add), then install retrieval-core@context-kit. It also
ships automatically with code-search.
rg, fd)ast-grep, semgrep)jq, yq, gron)git log -S/-G/-L)duckdb, sqlite-utils)tokei, scc)rga, pandoc)local-rag: rag index then rag query)obsidian bridge: graph/tags → rag query --allowlist). For authoring/Bases/Canvas use kepano/obsidian-skills.Lexical/structural/structured-data/history/metrics/docs live in code-search. Semantic and graph are available as the local-rag and obsidian plugins.
rag query --allowlist - (turbovec reranks only those).rg pins exact lines.rg -c) before reading full output.local-rag, obsidian) isn't installed, say so and
suggest installing it — don't assume its tools exist.rtk is installed, prefix the rtk-wrapped commands (rg/git/find/
diff) for compact output — it passes other tools through unchanged. (See
the code-search plugin's rtk reference.)npx claudepluginhub mbeacom/context-kit --plugin retrieval-coreGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.
Dispatches multiple subagents concurrently for independent tasks without shared state. Use when facing 2+ unrelated failures or subsystems that can be investigated in parallel.