Back to skill
Skillv1.0.0
ClawScan security
Reading · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 5:31 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This instruction-only reading advice skill is coherent with its stated purpose and does not request code execution, credentials, file access, persistence, or account authority.
- Guidance
- This appears safe to install as a lightweight reading-coaching skill. As with any advice skill, review whether its suggestions fit your goals, but the provided artifacts do not show security-relevant access or behavior.
Review Dimensions
- Purpose & Capability
- okThe skill provides reading recommendations, retention strategies, and guidance for matching reading methods to user goals, which aligns with its description.
- Instruction Scope
- okInstructions focus on asking context questions and offering reading-related advice; they do not redirect goals, require unsafe tool use, or ask the agent to override user intent.
- Install Mechanism
- okThere is no install spec and no code files; the artifact is instruction-only.
- Credentials
- okThe metadata declares no required binaries, environment variables, credentials, config paths, or OS-specific access.
- Persistence & Privilege
- okThe artifacts show no persistence mechanism, background behavior, elevated privileges, credential use, or account access.