Skip to content
Get packs

Hybrid Search Implementation

Guides the agent to combine vector and keyword search for better retrieval in RAG and search systems

Source Download skill
Repository
wshobson/agents
Stars
42,095
Licence
MIT
Files
2 files

What it does

This skill gives the agent a set of patterns for building search that uses both meaning-based (vector) search and exact-word (keyword) search, then merges the results. When installed, the agent can explain and recommend fusion methods such as Reciprocal Rank Fusion, weighted scoring, and reranking, and can point to a reference file with concrete templates and worked examples. It is guidance and templates rather than a tool that runs searches on its own.

Great for

  • Developers building retrieval-augmented generation (RAG) pipelines who find pure vector search misses exact names or codes
  • Engineers improving search over domain-specific vocabulary, product codes, or proper names
  • Teams deciding which fusion method (RRF, linear, cross-encoder, cascade) suits their latency and quality needs

Skip it if

  • Non-technical users who only want to search their own files; this skill assumes you are building a search system, not using one
  • Projects needing a verified, tested implementation, since the skill gives patterns and templates that must still be checked against your own data and stack

Try asking

  • “Help me add keyword search alongside our embedding search for a RAG chatbot over product manuals”
  • “Which fusion method should I use for a search engine where users type model numbers and descriptions?”
  • “Show me a template for reranking hybrid search results with a cross-encoder”
  • “Review my search code and tell me whether I am skipping keyword search or over-fetching”
Step-by-step guide: how to use Hybrid Search ImplementationInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    hybrid-search-implementation.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip hybrid-search-implementation.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip hybrid-search-implementation.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Name your vector database, keyword engine, and embedding model up front so the agent picks templates that fit your stack.
  • Ask the agent to test fusion weights on a sample of your own queries rather than accepting defaults.
  • Check the referenced templates in references/details.md against current library APIs before copying them into production code.

What's inside

  • references/details.md
  • SKILL.md

Redistributed unmodified from wshobson/agents under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.