create-agents-wizard

Guided creation of multiple OpenClaw agents and their workspace templates. Use when users ask to create agents in batch, configure new agents, or customize A...

Security Scan

The skill's declared purpose (batch scaffolding of OpenClaw agents) aligns with its instructions and included script; it performs only local file creation/copying and CLI calls and does not request credentials or external installs.

Purpose & Capability

Name/description match the actual behavior: collecting agent metadata, drafting six workspace files (or two in fast mode), and optionally running the OpenClaw CLI. No unrelated credentials, binaries, or config paths are requested.

Instruction Scope

SKILL.md confines actions to asking questions, drafting files, writing to the specified workspace, and optionally invoking 'openclaw' CLI. It does not instruct reading unrelated system files or sending data to external endpoints. The agent is instructed to write only after user confirmation.

Install Mechanism

No install spec; instruction-only skill with one small included bash helper. The script performs only filesystem operations (mkdir, cp, cat) and contains no network downloads or execution of fetched code.

Credentials

No environment variables, secrets, or external credentials are requested. The only external dependency referenced is the optional 'openclaw' CLI, which is coherent with the skill's purpose.

Persistence & Privilege

always is false and the skill does not attempt to modify other skills or system-wide agent settings. Its runtime actions are limited to creating or copying files in user-provided workspace paths.

Assessment

This skill appears coherent and low-risk, but before running or granting autonomous use: inspect any proposed workspace path to avoid accidental overwrites, review the generated drafts before confirming writes, and verify the scaffold.sh source if you plan to run it locally (the script can create/overwrite files when --force is used). Be aware the workflow may call the 'openclaw' CLI if available — ensure that CLI is trusted and configured as you expect. If you want to restrict writes, ask for a docs-only run (the SKILL.md supports skipping CLI changes).

Follow this workflow to create multiple agents with low user effort.

1) Confirm scope

1. Ask for total agent count N.
2. Ask each agent id (lowercase letters/digits/hyphens only).
3. Ask each workspace path (default ~/.openclaw/workspace-<id>).
4. Ask scaffold mode:
   • Standard (default): 6 files
     • AGENTS.md, SOUL.md, IDENTITY.md, BOOTSTRAP.md, USER.md, STYLE.md
   • Fast mode: 2 files only
     • AGENTS.md, SOUL.md
5. If user does not specify, use Standard by default.

2) Process one agent at a time

For each agent, collect info and draft files in fixed order.

1. Ask focused questions (use references/question-bank.md if available).
2. Produce an improved draft from user input.
3. Ask explicit confirmation: Confirm write / Revise.
4. Write only after confirmation; otherwise keep iterating.

Rules:

• Do not dump a long questionnaire; ask in small rounds (2–4 key questions each).
• Preserve user intent; do not change direction without consent.
• Respect selected mode (Standard 6 files or Fast 2 files).
• If user requests partial scaffolding, allow writing only the confirmed subset and continue/stop as requested.

3) Write files and create agent

After required files for one agent are confirmed:

1. If agent does not exist, create it:
   • openclaw agents add <id> --workspace <path>
2. Write approved files into workspace root.
   • Optional batch helper:
   • scripts/scaffold.sh --agent <id> --workspace <path> --from <approved_dir> --force
3. Optional identity setup:
   • openclaw agents set-identity --workspace <path> --from-identity

If user wants docs only, skip CLI config changes.

4) Final summary

When all agents are done, report:

• Created agents (id + workspace)
• Written file list per agent
• openclaw agents list output (if CLI available)
• Next-step suggestions (skills, routing, default models)

5) Quality bar

• AGENTS.md: operation rules, safety boundaries, heartbeat/external-action policy
• SOUL.md: values, working principles, boundaries, style
• IDENTITY.md: name, type, vibe, emoji, avatar convention
• BOOTSTRAP.md: first-conversation script and initialization steps
• USER.md: addressing, timezone, preferences, taboos, work/life context
• STYLE.md: tone, length preference, banned words, output format preference

6) Failure handling

• Conflicting user inputs: point out conflicts and offer 2 concrete revision options.
• Too-short answers: provide a minimal viable template, then request missing details.
• Scope change mid-way: reconfirm scope, then continue serial processing.
• Write failure: report exact path + error and provide retry command.