Back to skill
Skillv0.1.0
ClawScan security
ClawHub Intro Skill · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 8, 2026, 10:38 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- This is a minimal, instruction-only demonstration skill whose metadata and runtime instructions are consistent with its stated purpose.
- Guidance
- The skill is a harmless demo and internally consistent. Before installing, verify you trust the ClawHub registry and the clawhub CLI (installation/publish commands run via that tool). If you obtained this from an unknown source, prefer installing from the official registry and review the skill.json / SKILL.md first to ensure it matches the published registry entry.
Review Dimensions
- Purpose & Capability
- okName, description, and skill.json declare a demo/tutorial skill; the included files (SKILL.md, README.md, example.md, skill.json) and the described commands are exactly what you'd expect for a publishable example.
- Instruction Scope
- okSKILL.md only documents installation and publishing commands and lists package files. It does not instruct the agent to read unrelated files, access environment variables, transmit data to external endpoints, or perform privileged actions.
- Install Mechanism
- okNo install spec or code is included; this is instruction-only, so nothing is downloaded or installed by the skill itself. Installation would be performed by the external 'clawhub' CLI if the user runs it.
- Credentials
- okNo environment variables, credentials, or config paths are requested or referenced; the declared requirements are empty and match the SKILL.md content.
- Persistence & Privilege
- okSkill is not flagged as always:true and contains no install hooks or self-modifying steps. It is user-invocable only and does not request permanent presence or elevated agent privileges.