Skip to content
TakoAPI
Back to skills

search-memory

by trumppo

Coding Agents & IDEs
1.0kDownloads reported by clawskills.sh
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Local-first memory search and indexing for Openclaw.

Description

What This Skill Does

Indexes local memory files and runs keyword search with a recency boost. Covers MEMORY.md and all files under memory/, with an incremental cache to avoid full re-indexing on each run. Designed for Openclaw's local-first memory layout.

Keyword scoring combined with a 30/90-day recency boost surfaces recently updated notes first, avoiding manual scanning of every memory file.

When to Use It

  • Finding a past decision stored weeks ago in memory files
  • Locating notes about a specific project before resuming work
  • Searching for previously recorded config details or credentials
  • Retrieving context about a topic mid-conversation via slash command
  • Narrowing down relevant memory entries when the index has grown large

Example Workflow

INPUT

User asks: 'Find my notes about the auth middleware rewrite'

AGENT
  1. 1

    Run scripts/index-memory.py to build or update the incremental index

  2. 2

    Execute scripts/search-memory.py 'auth middleware rewrite' --top 5

  3. 3

    Apply keyword scoring and recency boost to rank results

  4. 4

    Return the top matching memory entries

OUTPUT

Top 5 ranked memory entries related to the auth middleware rewrite, ordered by relevance and recency

Installation

clawhub install trumppo/search-memory

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links