agentleFS
Sign inStart free
tumftumf/skills/openclaw-agent-creator/SKILL.md

openclaw-agent-creator

Skill4 starsChanged 7 months agoMITView on GitHub

Create and maintain OpenClaw agent workspaces using the "Memory as Documentation" layout. Use when you need to: (1) create a new OpenClaw workspace/profile, (2) generate bootstrap files (AGENTS.md, SOUL.md, USER.md, IDENTITY.md, TOOLS.md, MEMORY.md, memory/YYYY-MM-DD.md), (3) keep instructions/persona/memory separated and token-lean, or (4) set up a safe git backup of an OpenClaw workspace without committing secrets.

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a new workspace directory with templates:

    line 17
  2. Very specific:

    Prefer profiles: set OPENCLAW_PROFILE=myprofile when launching OpenClaw.

    line 35
  3. Very specific:

    If OpenClaw is installed, openclaw setup --workspace <path> can also seed missing bootstrap files without overwriting existing ones.

    line 39
  4. Distinctive:

    Do not store secrets in the workspace (even if backed up privately).

    line 56
  5. Distinctive:

    Keep bootstrap files short.

    line 57
  6. Very specific:

    Prefer writing durable notes into MEMORY.md and transient notes into memory/YYYY-MM-DD.md.

    line 58
  7. Distinctive:

    Create an OpenClaw workspace folder and seed bootstrap files.

    line 66

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

OpenClaw Agent Creator

This skill helps you create a new OpenClaw "agent persona" by generating a clean workspace and the standard bootstrap files.

Quick Start

  1. Create a new workspace directory with templates:
python3 "$SKILL_ROOT/scripts/init_workspace.py" \
  --workspace "$HOME/.openclaw/workspace-myprofile" \
  --agent-name "Claw" \
  --agent-vibe "sharp, concise, helpful" \
  --agent-emoji ":lobster:" \
  --user-name "tumf" \
  --user-language "Japanese" \
  --create-today-log

By default, this also creates a safe starter .gitignore to reduce the risk of committing secrets. Disable with --no-gitignore.

  1. Point OpenClaw at the workspace:
  • Prefer profiles: set OPENCLAW_PROFILE=myprofile when launching OpenClaw.
    • By default, this maps the workspace to ~/.openclaw/workspace-myprofile.
  • Or set the workspace path in ~/.openclaw/openclaw.json.

If OpenClaw is installed, openclaw setup --workspace <path> can also seed missing bootstrap files without overwriting existing ones.

Files and Responsibilities (keep them separated)

  • IDENTITY.md: stable identity metadata (name/vibe/emoji). Avoid tasks.
  • SOUL.md: stable behavioral rules (values, boundaries). Avoid temporary tasks.
  • USER.md: user preferences (language, formatting, expectations).
  • AGENTS.md: operating instructions (workflow, safety bar, memory policy).
  • TOOLS.md: local environment notes and "dangerous command" guardrails (guidance only).

Memory layer:

  • memory/YYYY-MM-DD.md: daily running log (append-only, messy is OK).
  • MEMORY.md: curated durable facts/decisions/lessons (keep small).

Safety / Hygiene

  • Do not store secrets in the workspace (even if backed up privately).
  • Keep bootstrap files short; OpenClaw may truncate them when injecting.
  • Prefer writing durable notes into MEMORY.md and transient notes into memory/YYYY-MM-DD.md.

Scripts

Note: Always call scripts via python3 "$SKILL_ROOT/scripts/<name>.py" ....

init_workspace.py

Create an OpenClaw workspace folder and seed bootstrap files. Defaults to safe behavior (no overwrites).

python3 "$SKILL_ROOT/scripts/init_workspace.py" --help

promote_keep.py

Promote KEEP: bullet lines from a daily log into MEMORY.md.

python3 "$SKILL_ROOT/scripts/promote_keep.py" \
  --daily "$HOME/.openclaw/workspace/memory/2026-02-19.md" \
  --memory "$HOME/.openclaw/workspace/MEMORY.md"

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free