bm25 plugins

21 tagged bm25, measured the same way as everything else here.

Browse within: claude-code-plugin 10

ownmem

01

grpcer/ownmem

Plugin Claude Code

Bundles 3 skills · 160 tokens together

Git-native project memory for AI coding agents, with deterministic local recall, evidence gates, and layered delivery that quotes, points, or abstains.

not rated 429 +7 changed 4d ago A tokens not measured original Apache-2.0

pluck

04

hunhee98/pluck

Plugin Claude Code

Bundles 1 MCP server

Fast and token-friendly code reading for AI coding agents. Symbol-aware MCP tools that replace cat/grep with 85% fewer tokens, sub-millisecond search, and a raw fallback that preserves cat/grep parity byte-for-byte.

not rated 42 4mo ago A tokens not measured original MIT

memini

06

eleboucher/memini

Plugin Claude Code

Bundles 10 skills, 9 commands · 547 tokens together

Persistent memory for AI coding agents via the memini memory service. 7 hooks, MCP tools, and skills that capture tool usage and inject prior context at session start and per prompt.

not rated 25 changed yesterday A tokens not measured AGPL-3.0

memtomem

08

memtomem/memtomem

Plugin Claude Code

Bundles 7 skills, 1 MCP server · 216 tokens together

Safe, explicit memory search and persistence workflows backed by the memtomem MCP server.

not rated 18 +3 changed 7d ago A tokens not measured original Apache-2.0

lethe

09

teimurjan/lethe

Plugin Claude Code

Bundles 2 skills, 1 hook · 118 tokens together

Persistent memory across Claude Code sessions. Markdown-first storage, hybrid BM25 + dense retrieval, clustered retrieval-induced forgetting, optional Haiku enrichment.

not rated 11 3mo ago A tokens not measured original MIT

ctx

10

pluto2060/CTX

Plugin Claude Code

Bundles 1 plugin

Cross-session memory for Claude Code — honest utility-measured context bootstrapper. G1 (decision timeline from git log) + G2 (docs/code search) with BM25 + optional vec-daemon semantic rerank. Visible failure warnings when components degrade.

not rated 9 4mo ago A tokens not measured original MIT

melchizedek

11

louis49/melchizedek

Plugin Claude Code

Bundles 1 MCP server

Persistent memory for Claude Code. Indexes your conversation transcripts and provides hybrid search (BM25 + vectors + reranker).

not rated 7 3mo ago A tokens not measured original MIT

cdeust/ai-architect-mcp-codebase

Plugin Claude Code

Bundles 3 skills, 24 commands, 18 agents, 6 hooks, 2 MCP servers · 597 tokens together

Rust MCP server that indexes ANY codebase into a LadybugDB property graph — source in 11 languages with full AST, plus docs/config/binaries as File nodes (.js→Imports, Markdown→References). Resolves imports and call chains, detects communities via Leiden, traces execution flows from entry points, and exposes 27 MCP…

not rated 7 +2 changed 17d ago A tokens not measured original MIT

clawness

13

fullymiddleaged/Clawness

Plugin Claude Code

Bundles 13 skills, 7 agents · 1,443 tokens together

Puts the right rules in context on every prompt, automatically — built for full-stack developers moving across frontend, backend, and SQL in one session, researchers working in bare or LaTeX-only directories, and teams who want adversarial review (security red/blue team, code critic) before merge. 222 rules across 30…

not rated 5 changed 12d ago A tokens not measured original MIT

agmem

14

albedoweb/agmem

Plugin Claude Code

Bundles 1 skill, 1 command, 1 hook · 93 tokens together

Persistent project memory for Claude Code — indexes your repo into local JSONL and auto-injects task-relevant context (BM25, no LLM calls, nothing leaves your machine).

not rated 2 2mo ago A tokens not measured original Apache-2.0

koshi

16

jsharma1105/Koshi

Plugin Claude Code

Bundles 5 agents · 340 tokens together

Koshi MCP — retrieval, context engineering, persistent memory, and quality scoring for AI agents. 100% offline.

not rated 2 1mo ago A tokens not measured original MIT

inmemory

17

Reikor-Arg/inmemory

Plugin Claude Code

Bundles 4 skills, 12 commands, 5 hooks · 384 tokens together

Verbatim recall across Claude Code sessions, plus a gate that stops oversized skills from flooding the prompt. Pure Node, no dependencies, no API calls.

not rated 1 2mo ago A tokens not measured original MIT

quartermaster

18

PranavNagrecha/Quartermaster

Plugin Claude Code

Bundles 1 skill · 38 tokens together

Offline, zero-dependency tool-router for MCP. Funnels many tools down to a ranked shortlist for a natural-language query — no embedding model, no network — so the model reads 8 tools instead of 200. Funnel advises, model decides.

not rated 0 3mo ago A tokens not measured original MIT