Knowledge retrieval pipeline review
Skill jukrap/ai-agent-playbook/skills/data/knowledge-retrieval-pipeline-review
Use when reviewing document ingestion, parsing, chunking, metadata, deduplication, embeddings, vector stores, retrieval quality, citations, or stale RAG/search indexes.From its SKILL.md
npx -y skills add jukrap/ai-agent-playbook --skill knowledge-retrieval-pipeline-reviewAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.2 KB, 203 tokens by cl100k_base, as published. Nobody here has run it
Knowledge Retrieval Pipeline Review
Use this as the primary data skill for document ingestion, retrieval pipelines, RAG/search indexes, and citation handoff.
Workflow
- Identify source documents, access control, parsing, chunking, metadata, deduplication, freshness, embedding/index boundary, and retrieval consumers.
- Check source provenance, private content handling, invalidation, stale chunks, permissions, evaluation set, citation behavior, and fallback states.
- Keep embedding/vector providers optional and disabled by default unless privacy, cost, and invalidation rules are explicit.
- Verify with source inventory, sampled chunks, retrieval tests, citation checks, access-control checks, and stale-index handling when possible.
Reference
Read references/document-ingestion-and-chunking.md for source loading, parsing, chunking, metadata, and dedup checks.
Read references/retrieval-evaluation-and-safety.md for retrieval evaluation, citations, permissions, stale indexes, and provider boundaries.
What ships with it: 2 files
2.4 KB alongside SKILL.md