hybrid-retriever

Runs hybrid retrieval with strict FTS-first policy, BM25 lexical ranking, vector merge, and top 3-5 reranking for token-efficient context selection.

Boost this agent

madhan230205@madhan230205

GitHub

From plugin

claude-token-reducer

444 skills4 agents1 command1 hook

Install

> /plugin marketplace add Madhan230205/token-reducer
> /plugin install claude-token-reducer@Madhan230205-claude-token-reducer

How it fires

How this agent gets triggered: by you, by Claude, or both.

Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.

Context preview

The summary Claude sees to decide when to auto-load this agent.

Runs hybrid retrieval with strict FTS-first policy, BM25 lexical ranking, vector merge, and top 3-5 reranking for token-efficient context selection.

Agent definition

hybrid-retriever.md

name: hybrid-retriever
description: Runs hybrid retrieval with strict FTS-first policy, BM25 lexical ranking, vector merge, and top 3-5 reranking for token-efficient context selection.
model: sonnet
color: blue
tools: Read, Glob, Grep

You are a retrieval specialist.

1. Query FTS first. 2. Rank FTS hits with BM25. 3. Run vector retrieval adaptively (fallback default) and merge with FTS results. 4. Re-rank by BM25 lexical signal, semantic signal, and query overlap. 5. Return 3-5 chunks maximum.

Read it on GitHub ↗

Ships withclaude-token-reducer

⚡ Cut Claude token usage by 90%+ — free, open-source, local-first context compression for Claude Code. Hybrid RAG (BM25 + ONNX vectors), AST chunking, reranking. No API needed.

Get the whole plugin

Stats

Python TypeScript JavaScript HTML Bash

+7 tags

Rust Java Ruby SQLite Git GitHub Anthropic

Stars

Forks

Maintained

Maintenance

Python

Language

MIT