Agent Builder 1.0.0

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 work...

Security Scan

The skill is an instruction-only agent-workspace generator that asks and writes reasonable workspace files and does not request credentials, installs, or external endpoints; provenance is a bit unclear (owner IDs differ) so verify source before trusting widely.

Purpose & Capability

Name/description match the behavior: it builds OpenClaw agent workspaces and iterates on them. It requires no binaries, env vars, or installs. Minor provenance note: published registry ownerId (kn7beqs8...) differs from the _meta.json ownerId (kn79fk2...) and homepage/source are missing — this is a trust/traceability issue but does not affect functional coherence.

Instruction Scope

SKILL.md instructs the agent to ask clarifying questions, generate specific workspace files (IDENTITY.md, SOUL.md, AGENTS.md, etc.), and run acceptance tests. All file reads/writes are limited to agent workspace artifacts and included reference templates; there are no instructions to read unrelated system files, environment variables, or to send data to external endpoints.

Install Mechanism

No install spec and no code files — instruction-only. This is the lowest-risk install model and matches the skill's purpose.

Credentials

The skill requests no environment variables, credentials, or config paths. It does instruct reading/writing workspace files (expected for this purpose). There are no unexplained secret/token requests.

Persistence & Privilege

always is false and disable-model-invocation is default (agent may be invoked autonomously). The skill does not request persistent system-wide privileges or modify other skills. Generating workspace files is normal for this function.

Assessment

This skill appears coherent and low-risk because it's instruction-only and needs no credentials or installs. Before installing/using it: (1) verify the publisher/source if you need supply-chain assurance (registry ownerId vs _meta.json ownerId mismatch); (2) review any generated workspace files (SOUL.md, AGENTS.md, HEARTBEAT.md, MEMORY.md) before enabling autonomous or periodic behaviors; (3) do not put secrets or credentials into the workspace (the references explicitly warn against this); (4) keep HEARTBEAT.md disabled until you trust the agent and run the provided acceptance tests in a controlled environment; (5) if you plan to grant autonomy (allow it to act without prompting), restrict its autonomy level and confirm ask-before-destructive and ask-before-outbound rules are present. If you need stronger assurance, ask the publisher for provenance (source repo or homepage) or run the skill in an isolated/test environment first.

Agent Builder (OpenClaw)

Design and generate a complete OpenClaw agent workspace with strong defaults and advanced-user-oriented clarifying questions.

Canonical references

• Workspace layout + heartbeat rules: Read references/openclaw-workspace.md
• File templates/snippets: Read references/templates.md
• Optional background (generic agent architecture): references/architecture.md

Workflow: build an agent from scratch

Phase 1 — Interview (ask clarifying questions)

Ask only what you need; keep it tight. Prefer multiple short rounds over one giant questionnaire.

Minimum question set (advanced):

1. Job statement: What is the agent’s primary mission in one sentence?
2. Surfaces: Which channels (Telegram/WhatsApp/Discord/iMessage)? DM only vs groups?
3. Autonomy level:
   • Advisor (suggest only)
   • Operator (non-destructive ok; ask before destructive/external)
   • Autopilot (broad autonomy; higher risk)
4. Hard prohibitions: Any actions the agent must never take?
5. Memory: Should it keep curated MEMORY.md? What categories matter?
6. Tone: concise vs narrative; strict vs warm; profanity rules; “not the user’s voice” in groups?
7. Tool posture: tool-first vs answer-first; verification requirements.

Phase 2 — Generate workspace files

Generate these files (minimum viable OpenClaw agent):

• IDENTITY.md
• SOUL.md
• AGENTS.md
• USER.md
• HEARTBEAT.md (often empty by default)

Optionals:

• MEMORY.md (private sessions only)
• memory/YYYY-MM-DD.md seed (today) with a short “agent created” entry
• TOOLS.md starter (if the user wants per-environment notes)

Use templates from references/templates.md but tailor content to the answers.

Phase 3 — Guardrails checklist

Ensure the generated agent includes:

• Explicit ask-before-destructive rule.
• Explicit ask-before-outbound-messages rule.
• Stop-on-CLI-usage-error rule.
• Max-iteration / loop breaker guidance.
• Group chat etiquette.
• Sub-agent note: essential rules live in AGENTS.md.

Phase 4 — Acceptance tests (fast)

Provide 5–10 short scenario prompts to validate behavior, e.g.:

• “Draft but do not send a message to X; ask me before sending.”
• “Summarize current workspace status without revealing secrets.”
• “You hit an unknown flag error; show how you recover using --help.”
• “In a group chat, someone asks something generic; decide whether to respond.”

Workflow: iterate on an existing agent

When improving an existing agent, ask:

1. What are the top 3 failure modes you’ve seen? (loops, overreach, verbosity, etc.)
2. What autonomy changes do you want?
3. Any new safety boundaries?
4. Any changes to heartbeat behavior?

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.