Search History
Documentssearch-historySearch History is an Agent Skill for documents published in the majiayu000/claude-skill-registry repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Search Claude conversation history from JSONL files. Use when looking for previous discussions, past decisions, code solutions, or context from earlier conversations.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/majiayu000/claude-skill-registry/blob/HEAD/skills/data/search-history/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/search-history, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
Conversation History Search
Search through Claude Code conversation history to find relevant past discussions.
When to Use
- Finding previous discussions about a topic
- Recalling past decisions or solutions
- Getting context from earlier conversations
- Looking up how something was implemented before
How to Search
Run the search script from the services directory:
cd ~/.claude/services
python conversation-search.py "search term"
Search Options
| Option | Description |
|---|---|
--list-sessions, -l | List all sessions with metadata |
--session ID, -s | Filter to specific session (partial match) |
--context N, -c | Show N messages before/after (default: 3) |
--max N, -m | Maximum results (default: 20) |
--verbose, -v | Show context messages |
--full, -f | Show full message content |
Examples
# List all available sessions
python conversation-search.py --list-sessions
# Basic search
python conversation-search.py "bulk upload"
# Regex search
python conversation-search.py "context.*percent|usage.*%"
# Search specific session with full output
python conversation-search.py "statusline" --session 6cabef43 --full
# Get more context around matches
python conversation-search.py "error" --context 5 --verbose --max 3
Output Modes
- Default: Preview snippet (50 chars before, 100 after match)
- Verbose (
-v): Adds surrounding message context - Full (
-f): Complete message content
Token Efficiency
The search runs externally, so searching costs 0 tokens. Only the returned results consume context.