m3-search

Solid

Hybrid memory search — FTS5 + semantic vector + MMR diversity rerank.

AI & Automation 16 stars 3 forks Updated today Apache-2.0

Install

View on GitHub

Quality Score: 77/100

Stars 20%
41
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
35
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# M3 Search ## When to Use Use this skill when the user wants to search their memories, find information they previously saved, check past conventions or decisions, or resolve queries involving prior context. ## Instructions Call the `m3:memory_search` MCP tool with `query="$ARGUMENTS"` and `k=8`. Render the results as a numbered list: ``` 1. <title> — <type> — score=<float> <first 200 chars of content> id: <uuid-prefix> ``` If the user wants to drill into one, suggest `/m3:get <id-prefix>`. If results look stale or contradictory, suggest `/m3:forget <id>` for stale entries or `/m3:save <correct-fact>` to add the new state (contradiction detection will supersede automatically).

Details

Author
skynetcmd
Repository
skynetcmd/m3-memory
Created
5 months ago
Last Updated
today
Language
Python
License
Apache-2.0

Integrates with

Bundled in these plugins

Similar Skills

Semantically similar based on skill content — not just same category