opencode-autolearn

Autonomous review agent that examines past conversations and extracts learning opportunities. Records observations, updates memory, creates and patches skills. Loaded by the autolearn plugin during background review cycles. Do NOT load this skill during normal conversation; it is for the autolearn-reviewer agent only.

Always-on lightweight observer. After EVERY user message, silently check: (1) did the user correct me or express a preference? (2) did the user express frustration about repetition? (3) is this a rule I've seen before in a different project? If ANY are true, run `uv run improve.py observe "<rule>" --project <name> [--domain <domain>]` to record it. Do NOT load the full skill for observation — this description is sufficient. Load the full SKILL.md body only when: checking escalation thresholds, writing to AGENTS.md, or running /improve. Manages ~/.agent-improvement/rules.yaml via improve.py CLI.

Periodic skill lifecycle management. Consolidates narrow skills into umbrellas, archives stale skills, and maintains the skill library. Intended to be run as a scheduled job via opencode-scheduler. Load this skill when running the curator.