Bundle Pluginstructural

OpenClaw Skill Creator

Bundle plugin: guide for creating or updating reusable OpenClaw skills

@aisa/skill-creator · runtime id @aisa/skill-creator

Install

openclaw bundles install clawhub:@aisa/skill-creator

Latest Release

Version 0.1.1

Download zip

Compatibility

{
  "builtWithOpenClawVersion": "0.1.0"
}

Capabilities

{
  "bundleFormat": "generic",
  "capabilityTags": [
    "bundle-only",
    "format:generic",
    "host:openclaw"
  ],
  "executesCode": false,
  "hostTargets": [
    "openclaw"
  ],
  "runtimeId": "@aisa/skill-creator"
}

Security Scan

VirusTotal

Pending

View report →

OpenClaw

Benign

high confidence

✓

Purpose & Capability

The name/description (skill-creator) match the included files: SKILL.md guidance and two helper scripts (init_skill.py, quick_validate.py). The scripts and references are appropriate for scaffolding and validating OpenClaw skills; nothing requested is unrelated to that purpose.

ℹ

Instruction Scope

SKILL.md instructs running the bundled Python scripts and using the platform's message tool. The scripts read and write files under /home/ubuntu/skills (create SKILL.md, example resources, and validate SKILL.md frontmatter). They do not reference network endpoints or secrets. One implementation detail to note: init_skill.py joins the provided skill name with the base path but does not sanitize or canonicalize the user-supplied skill-name argument, so an adversarial or malformed argument containing path traversal components (e.g., '../') could cause directories to be created outside the intended base. Review usage and validate inputs before running.

✓

Install Mechanism

There is no external install spec or network download; this is an instruction-and-scripts bundle. Files are included in the package and no remote code is fetched at install time.

✓

Credentials

The skill declares no required environment variables, tokens, or credentials and the scripts do not read environment secrets. Nothing requests unrelated secrets or config paths.

ℹ

Persistence & Privilege

The skill does not request persistent platform privileges (always:false). The scripts create files under /home/ubuntu/skills and will persist scaffolded skill directories on disk. As noted above, lack of sanitization for the skill-name argument could allow creating files outside the intended directory if called with crafted input; run scripts under a non-privileged account and review arguments before invoking.

Assessment

This package appears to be what it claims: a local skill-authoring guide with two helper scripts. Before running anything: 1) inspect init_skill.py and quick_validate.py yourself (they are short and included) to confirm behavior; 2) run them as a non-root, unprivileged user so created files can't affect the whole system; 3) be cautious when supplying the <skill-name> argument — avoid special characters or path separators (the initializer does not canonicalize/sanitize the name and could create directories outside /home/ubuntu/skills if given '../' components); 4) note there is no network access or credential usage in the code, but if you plan to modify or run generated scripts, re-check for any added network calls or secrets handling. If you need stronger assurance, run the scripts in an isolated test environment (container or VM) first.

Verification

{
  "scanStatus": "pending",
  "scope": "artifact-only",
  "summary": "Validated package structure and extracted metadata.",
  "tier": "structural"
}