Back to skill
Skillv1.0.0
ClawScan security
my skill · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 3, 2026, 3:07 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- This is an instruction-only skill for reviewing and improving SKILL.md files; its declared behavior, files, and requirements are coherent and it does not request unrelated credentials, installs, or network access.
- Guidance
- This skill appears coherent and low-risk: it only analyzes SKILL.md content and provides remediation guidance. Before using it, avoid giving it paths to system or secret files (e.g., ~/.aws/credentials, /etc/*); instead paste or provide only the SKILL.md content you want reviewed. Back up any original SKILL.md before applying automated fixes. If you plan to let agents invoke this skill autonomously, consider restricting which files or directories the agent can access to prevent accidental reading of sensitive files.
Review Dimensions
- Purpose & Capability
- okName/description match the included assets (skill-reviewer and skill-standard). The files and instructions are precisely what a SKILL.md reviewer/formatter would need; there are no unrelated environment variables, binaries, or install steps.
- Instruction Scope
- noteRuntime instructions are limited to parsing or reading the target SKILL.md (either via provided path or pasted content) and applying the review rubric. This is appropriate, but the reviewer does rely on reading user-supplied file paths—avoid passing system or secret file paths to the skill. The conditional loading of skill-standard only on C/D results is explicit and reasonable.
- Install Mechanism
- okNo install spec or code files beyond Markdown are included (instruction-only). Nothing is written to disk or downloaded by the skill itself as part of installation.
- Credentials
- okNo environment variables, credentials, or config paths are requested. The skill does not require secrets or unrelated service tokens.
- Persistence & Privilege
- okalways is false and the skill does not request persistent or cross-skill configuration changes. Model invocation is allowed (platform default) but that is appropriate for an autonomous reviewer; no excessive privileges are requested.