Back to skill
Skillv0.1.0
ClawScan security
暗号对接（猛如虎→小松鼠） · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 8:28 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is a simple instruction-only codeword skill that replies with a fixed phrase and does not request code execution, credentials, files, or system access.
- Guidance
- This skill appears safe for its stated purpose. Install it only if you want the agent to give the fixed reply whenever the exact phrase “一顿操作猛如虎” appears.
Review Dimensions
- Purpose & Capability
- okThe skill purpose is narrow and matches the described behavior: respond with a fixed line when the exact codeword phrase appears.
- Instruction Scope
- okThe response instruction is tightly scoped to the full trigger phrase and does not ask the agent to ignore unrelated user goals outside that trigger.
- Install Mechanism
- okNo install spec, binaries, environment variables, credentials, or code files are present.
- Credentials
- okThe skill does not request local file access, network access, account access, or other environmental authority.
- Persistence & Privilege
- okThere is no evidence of persistence, background behavior, privilege escalation, stored memory, or credential use.