All 94 skillsWeb and data3 of 17

stryx:rag

When an app must search its own documents and answer from them.

/stryx:rag

Inside its piece: 12 specks, one for each file on its shelf. Its sixteen group-mates glow on the open book behind it. The light splits toward teal, the band of Web and data.

In numbers

counted 2026-10-12 in the installed set
  • 9reference files
  • 1helper scripts
  • 12files in the skill
  • 124KB on disk
  • 2chapters
  • 1agent works by this book

Use it when

in the skill’s own words

When building retrieval-augmented generation (RAG) or a search engine into an app, script or agent

    • "build a RAG pipeline"
    • "chat with our docs"
    • "add semantic / vector search"
    • "ground answers in live web data"
    • "add web search to my LLM or agent"
    • "ingest pages or files into a vector DB / knowledge base"
  • Choosing live retrieval vs ingestion
  • Chunking
  • Embeddings incl. the local stryx-embed (nomic-embed-text)
    • pgvector
    • sqlite-vec
    • Chroma
    • Qdrant or a sql.js vector table
  • Hybrid keyword + vector search, RRF, reranking
  • Answers that must cite sources ([n] citations, Claude search_result blocks)
  • A stale or poisoned index (block pages retrieved, wrong chunks, invented citations)
  • Measuring retrieval (hit@k, MRR, grounding)

Not for

another skill does it

On its shelf

12 files, 124 KB
  • references/9
  • SKILL.md1
  • evals/1
  • scripts/1

Inside the book

2 chapters, by title only
  1. The procedure
  2. Scripts

Its helper

1 agent works by this book

Works with

the skills it names most

Claude loads it by itself when the moment comes; by hand, type the line at the top. It comes with the whole set: Installs all 94 skills, 32 agents and the brain program as one set. The skill’s own text is private source and is not shown here.