Back to skill
Skillv1.0.0
ClawScan security
执行申请助手 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 7:33 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This instruction-only legal procedure helper appears coherent with its stated purpose and does not request code execution, credentials, installation, persistence, or system access.
- Guidance
- This appears safe from an agentic-security perspective. Treat it as procedural information rather than legal advice, verify deadlines and local court requirements, consult a qualified lawyer for complex matters, and avoid sharing unnecessary personal, identity, financial, or debtor asset details.
Review Dimensions
- Purpose & Capability
- okThe artifacts consistently provide civil enforcement application templates, document checklists, process guidance, and procedural disclaimers aligned with the skill description.
- Instruction Scope
- okThe instructions are bounded to drafting and explaining civil enforcement applications and do not contain goal overrides, hidden tool-use requirements, or autonomous action instructions.
- Install Mechanism
- okThere is no install spec, no required binaries, no environment variables, and no code files; this is an instruction-only skill.
- Credentials
- okNo filesystem, network, OS, credential, or external-tool capabilities are declared or evidenced by the artifacts.
- Persistence & Privilege
- okThe artifacts show no persistence, background execution, privileged access, account access, or long-running agent behavior.