Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Searches OpenClaw session transcripts stored as JSONL files to retrieve past conversations. Finds specific topics, decisions, or code from previous sessions and helps resume dropped threads.
Keyword search across raw JSONL transcripts reaches content that memory files and semantic indexes miss.
When to use it
- Finding a code snippet discussed in a session two weeks ago
- Resuming a debugging thread that was dropped mid-conversation
- Recalling a decision reached in a past session
- Locating an error message from a previous troubleshooting conversation
- Cross-referencing session content against current memory files
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: find that conversation where we discussed the gclid pipeline error
- 1Run search_sessions.py with 'gclid pipeline error' as the query
- 2Review matching results and identify the relevant session key
- 3Pull full context using sessions_history tool with the matched sessionKey
- 4Fall back to reading the JSONL file directly if sessions_history is unavailable
- 5Summarize the relevant discussion from that session
A summary of the gclid pipeline error discussion, including what was tried and what conclusions were reached