Back to skill
Skillv1.1.1
ClawScan security
[Nyx Archive] Coding Philosophy · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 5:59 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is an instruction-only coding philosophy/refactoring skill with no code, install steps, credentials, or suspicious behavior shown in the artifacts.
- Guidance
- This appears safe to install as a writing/refactoring guidance skill. As with any coding advice, review code changes before applying them, especially cleanup or deletion recommendations.
Review Dimensions
- Purpose & Capability
- okThe visible SKILL.md content aligns with the stated purpose: coding philosophy, prototyping, and refactoring guidance.
- Instruction Scope
- okThe instructions are advisory and refactoring-focused, including cautious guidance such as auditing before cutting and committing before removing dead code.
- Install Mechanism
- okThere is no install spec, no required binaries, no required environment variables, and no code files.
- Credentials
- okThe skill requests no local system access, network access, credentials, or external tools beyond ordinary instruction use.
- Persistence & Privilege
- okThe artifacts show no persistence mechanisms, background behavior, credential handling, or privilege escalation.