Back to skill
Skillv0.1.1
ClawScan security
聊天助理 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 8:15 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This skill appears to be a simple chat-reply helper with no evidence of credential use, network access, persistence, or account-changing behavior.
- Guidance
- This looks proportionate for a chat-advice skill. Users should still treat its relationship interpretations as suggestions rather than facts, but there is no artifact-backed security concern requiring review before installation.
Review Dimensions
- Purpose & Capability
- okThe SKILL.md description and included scripts consistently focus on generating dating/chat reply suggestions, opening lines, and conversation analysis.
- Instruction Scope
- okInstructions are user-directed and limited to providing suggested wording, analysis, and warnings; they do not ask the agent to override user intent or take actions outside the chat-advice purpose.
- Install Mechanism
- okThere is no install spec, no declared dependency installation, and package.json lists no dependencies; included scripts are local CLI-style helpers and are not presented as automatically executed.
- Credentials
- okNo required environment variables, credentials, binaries, config paths, network calls, file reads, or file writes are evidenced in the artifacts.
- Persistence & Privilege
- okThe artifacts show no persistence, background execution, privilege escalation, account access, or long-running behavior.