From openclaw
Build high-performing OpenClaw agents end-to-end. Use when you want to design a new agent (persona + operating rules) and generate the required OpenClaw workspace files (SOUL.md, IDENTITY.md, AGENTS.md, USER.md, HEARTBEAT.md, optional MEMORY.md + memory/YYYY-MM-DD.md). Also use to iterate on an existing agent's behavior, guardrails, autonomy model, heartbeat plan, and skill roster.
How this skill is triggered — by the user, by Claude, or both
Slash command
/openclaw:openclaw-agent-builderThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Design and generate a complete **OpenClaw agent workspace** with strong defaults and advanced-user-oriented clarifying questions.
Design and generate a complete OpenClaw agent workspace with strong defaults and advanced-user-oriented clarifying questions.
references/openclaw-workspace.mdreferences/templates.mdreferences/architecture.mdAsk only what you need; keep it tight. Prefer multiple short rounds over one giant questionnaire.
Minimum question set (advanced):
MEMORY.md? What categories matter?Generate these files (minimum viable OpenClaw agent):
IDENTITY.mdSOUL.mdAGENTS.mdUSER.mdHEARTBEAT.md (often empty by default)Optionals:
MEMORY.md (private sessions only)memory/YYYY-MM-DD.md seed (today) with a short "agent created" entryTOOLS.md starter (if the user wants per-environment notes)For memory setup details (daily logs, MEMORY.md curation, shared memory, SCRIBE compression), see
openclaw-memoryskill. Use/oc-memory initto scaffold the memory structure.
Use templates from references/templates.md but tailor content to the answers.
If the agent's autonomy level is Operator or Autopilot, generate these additional configurations based on the Phase 1 autonomy model answers:
## Autonomous Behavior section in SOUL.md — Explicit idle triggers, proactive check rules, frequency, and boundaries (see openclaw-autonomy-audit skill, Pattern A)- [ ] checklist items based on the agent's job statement and idle behavior preferencesopenclaw cron add commands for scheduled tasks (health checks, inbox scans, summaries) matching the stated frequency## Cost Controls section in AGENTS.md — Max turns per cycle, max retries, stop conditions, circuit breaker rules, daily budget (see openclaw-autonomy-audit skill, Pattern E)Ensure the generated agent includes:
AGENTS.md.Provide 5–10 short scenario prompts to validate behavior, e.g.:
When improving an existing agent, ask:
/oc-autonomy --agent <id> for a scored assessment.Then propose targeted diffs to:
SOUL.md (persona/tone/boundaries)AGENTS.md (operating rules + memory + delegation)HEARTBEAT.md (small checklist)Keep changes minimal and surgical.
npx claudepluginhub theophiluschinomona/claude-openclaw-plugin --plugin openclawCreates Claude Code agents from scratch or by adapting templates. Guides requirements gathering, template selection, and file generation following Anthropic best practices (v2.1.63+).
Guides creation and configuration of autonomous agents for Claude Code plugins, covering frontmatter, triggering descriptions, system prompts, tools, teams, permissions, and best practices.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.