Recall

When a user asks "what did we do about X" or "have we seen Y before," this skill searches agentmemory using hybrid BM25, vector, and graph search to surface past observations, decisions, and code snippets across sessions. It groups results by session and surfaces high-importance items first, which matters because not all context is equally useful. The discipline here is real: if the tool returns nothing, you suggest alternative search terms instead of hallucinating memories. That constraint, never fabricate an observation or session ID, makes this trustworthy for long-running projects where you actually need to know whether something happened or you're just guessing.

skills.sh installs: 14.5K
GitHub stars: 30.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add rohitg00/agentmemory --skill recall --agent claude-code

Installs into .claude/skills of the current project.

Files

The user wants to recall past context about: $ARGUMENTS

Quick start

memory_smart_search { "query": "jwt refresh token rotation", "limit": 10 }

Expected output:

2 results across 2 sessions.
[importance 8] decision · "Rotate refresh tokens on every use" (session 7f3a9c21)
[importance 5] code · "limit.ts counts per-IP" (session b21d004e)