Documentation Index Fetch the complete documentation index at: https://auto-sop.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
learn-now Run the learner immediately instead of waiting for the hourly scheduler. By default, LLM analysis is enabled (uses your Claude Max subscription at $0 cost).
auto-sop learn-now [options]
Options Option Description --dry-runShow what would change without writing --offlineDisable LLM mode (rule-based detectors only) --force-llmForce LLM analysis even if no new turns --recomputeForce full metrics recomputation from all turns --limit <n>Limit recap entries shown
Examples Run Learner Now
Preview Without Writing
Rule-Based Only (No LLM)
Force Full Recomputation
How Learning Works
Scans captured interactions since last runRule-based detectors identify patterns requiring N>=3 evidence instancesLLM analysis (via claude -p) finds higher-level patterns across sessionsValidates candidate directives against existing ones (no duplicates)Writes new directives to the managed section of CLAUDE.md
Use --dry-run to preview what directives would be added before committing them.
recap Show learner recap log entries. Each entry records when the learner ran, what it found, and what directives were written.
Options Option Description --limit <n>Show last N entries (default: 10) --tailFollow recap log for new entries --followAlias for --tail --runRun the learner now (LLM mode ON by default) --dry-runDry-run mode (requires --run) --offlineDisable LLM mode (requires --run)
Examples View Recent Recaps
Show Last 5 Entries
Follow Live
Run + Show Results
--run is a convenience shortcut that combines learn-now with viewing the recap log. It’s equivalent to running auto-sop learn-now followed by auto-sop recap.
candidates Show, prune, or clear LLM pattern candidates. Candidates are potential directives identified by the LLM that haven’t yet accumulated enough evidence to be promoted.
auto-sop candidates [options]
Options Option Description --project <path>Project root (default: current directory) --pruneRemove stale candidates older than 30 days --clearClear all candidates (fresh start)
Examples View Current Candidates
Prune Stale Candidates
Clear All Candidates
--clear removes all candidate patterns permanently. Use this if you want the LLM to start fresh without prior hypotheses.
