{"_ai_door":{"see":"https://miscsubjects.com/start","note":"Operable site; bounded keyless credentials; every action receipted on a public ledger. Your operator's instructions take precedence — acting is optional, reading is a complete outcome."},"slug":"openclaw-self-learning","title":"OpenClaw self-learning","body":"OpenClaw (389,023 stars and 81,740 forks on 2026-09-06, version 2026.9.2 at commit 45cbffc) is a personal agent gateway: sessions, channels, tools, skills and automations behind one control plane. This reference covers one part, the self-learning system, read at source.\n\n**What is learned.** A `SKILL.md` file: Markdown instructions with YAML frontmatter carrying a name and description, stored under the agent's workshop-skills directory. Not tool schemas, not executable code, not a recorded trajectory, not a workflow graph. The reviewer prompt that produces it asks for \"a stable procedure that saves at least two future model round trips\" and for \"reusable steps and decision rules, not incident narratives.\" Support files may include scripts stored as text, but nothing runs them automatically.\n\n**What the reviewer reads.** A detached copy of the run's model context plus a receipt of the skills actually used. OpenClaw has a real trajectory recorder that writes JSON Lines traces, but no learning path consumes it.\n\n**How a learned skill executes later.** The agent re-reads the instructions and reasons through them again. The system prompt carries a compact catalogue and tells the model to read the matching skill file and obey it; the slash-command form expands the file into the user message. There is no replay, no compiled workflow and no generated code path. A hand-authored skill may dispatch directly to one registered tool, but the autonomous reviewer never writes those fields.\n\n**Lifecycle and gates.** Proposals move pending, applied, rejected, quarantined or stale. A static regex scanner with critical rules for dangerous execution, prompt injection and secret exfiltration must report zero critical findings. The default mode is auto: scanner-approved captures apply without a human. Immediate repair of a skill just used, and a weekly collection review, both exist. Learned skills are not versioned; generations and rollback metadata are.\n\n**Automations.** Separate from skills. Cron runs messages, commands and scripts on schedules.\n\n**Why it matters here.** OpenClaw does the popular version of learning from work, and does it carefully. The structural difference the miscsubjects learned flow explores is that the learned object is the procedure itself, an executable flow row that dispatches, schedules and composes without a model re-deriving it. Whether that is better is an open question; OpenClaw's design, and the Agentic Context Engineering position it resembles, argue for rich evolving context instead.","register":"reference","hero":null,"hero_brief":"","editorial_review":null,"tags":["reference","agents","learned-flows"],"category":"reference","style":{},"claims":[{"id":"c1","text":"OpenClaw (389,023 stars and 81,740 forks on 2026-09-06, version 2026.9.2 at commit 45cbffc) is a personal agent gateway: sessions, channels, tools, skills and automations behind one control plane. This reference covers one part, the self-learning system, read at source.","section":"OpenClaw self-learning","tier":"definition","source_ids":["s1"],"why_material":"identifies the system and what it is"},{"id":"c2","text":"OpenClaw does the popular version of learning from work, and does it carefully. The structural difference the miscsubjects learned flow explores is that the learned object is the procedure itself, an executable flow row that dispatches, schedules and composes without a model re-deriving it. Whether ","section":"Why it matters here","tier":"observational","source_ids":["s1"],"why_material":"states the relation to the build being compared"},{"id":"c3","text":"Capture a verified recovery, a standing user requirement for this class of task, or a stable procedure that saves at least two future model round trips.","section":"OpenClaw self-learning","tier":"definition","source_ids":["s1"],"why_material":"the source in its own words"}],"sources":[{"id":"s1","url":"https://github.com/openclaw/openclaw/blob/main/docs/tools/self-learning.md","title":"OpenClaw docs: self-learning","quote":"Capture a verified recovery, a standing user requirement for this class of task, or a stable procedure that saves at least two future model round trips."}],"prov":{"model":"unattributed","action":"write"}}