Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.selftune.dev/llms.txt

Use this file to discover all available pages before exploring further.

​
Usage

selftune sync [options]
Scans configured agent platforms (Claude Code, Codex, OpenCode, OpenClaw, Pi) and ingests all new session data into the selftune database.

​
What it scans

PlatformSourceData
Claude CodeClaude Code hooksPrompts, tool calls, results
CodexCodex logsAgent interactions
OpenCodeOpenCode data directorySession events
OpenClawOpenClaw agents directoryAgent execution traces
PiPi sessions directoryAgent messages and tool calls

​
Options

--since <date>                   Only sync sessions after this date (ISO 8601)
--dry-run                        Show summary without writing to SQLite or files
--force                          Rescan all sessions, ignore markers
--no-claude                      Skip Claude transcript replay
--no-codex                       Skip Codex rollout ingest
--no-opencode                    Skip OpenCode ingest
--no-openclaw                    Skip OpenClaw ingest
--no-pi                          Skip Pi ingest
--no-repair                      Skip rebuilding repaired skill-usage data
--json                           Output results as JSON
--projects-dir <dir>             Claude transcript directory (default: ~/.claude/projects)
--codex-home <dir>               Codex home directory (default: ~/.codex)
--opencode-data-dir <dir>        OpenCode data directory
--openclaw-agents-dir <dir>      OpenClaw agents directory
--pi-sessions-dir <dir>          Pi sessions directory (default: ~/.pi/agent/sessions)
--skill-log <path>               Raw skill usage log path
--repaired-skill-log <path>      Repaired overlay log path
--repaired-sessions-marker <p>   Repaired session marker path

​
Examples

Sync all platforms: 
selftune sync
Sync only sessions from the last 24 hours: 
selftune sync --since 2026-04-07
Sync with custom Pi sessions directory: 
selftune sync --pi-sessions-dir ~/my-pi-data/sessions
Rescan everything and re-ingest all sessions: 
selftune sync --force

​
Output

Returns a JSON summary with sync results: 
{
  "status": "success",
  "metrics": {
    "total_synced": 42,
    "claude_synced": 15,
    "codex_synced": 8,
    "opencode_synced": 10,
    "openclaw_synced": 5,
    "pi_synced": 4
  },
  "elapsedMs": 2341
}

​
Next steps

After syncing, analyze your data: 
selftune status          # View sync statistics
selftune grade           # Grade agent performance
selftune eval <name>     # Evaluate a specific skill