Skillv1.0.0

ClawScan security

职业发展助手 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 27, 2026, 1:42 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill is internally coherent and matches its stated purpose (local knowledge-base career adviser with an optional commercial API mode); no hidden/networked behavior or excessive privileges were found, though the commercial-API metadata/implementation details should be clarified before enabling it.
Guidance: This skill appears benign and mainly uses local reference files and a simple Python recommender. Before enabling commercial mode or supplying CAREER_API_URL / CAREER_API_KEY: 1) ask the publisher how/where the API config is used (the included script does not call external APIs); 2) only provide credentials to trusted endpoints and avoid giving highly sensitive keys; 3) prefer testing the skill in an isolated environment first; 4) request that the registry metadata be updated to list any required env vars and a homepage/source so you can verify the author and license. If you plan to use the commercial API integration, get details on what user data will be sent to the external endpoint and how it is protected.

Review Dimensions

Purpose & Capability: noteThe skill's name/description match the provided files: an offline knowledge base (references/*.md) and a local recommender script. SKILL.md also advertises a '商用模式' using CAREER_API_URL / CAREER_API_KEY, but the repository metadata does not declare required env vars and the included script (career-recommender.py) contains no code to call external APIs — this mismatch is an implementation/metadata inconsistency that should be clarified.
Instruction Scope: okRuntime instructions are narrowly scoped: identify user type, look up the built-in knowledge base, optionally call a commercial API if configured, and fall back to local content on API failure. The instructions do not ask the agent to read arbitrary host files or credentials, nor to exfiltrate data to unexpected endpoints. Note: SKILL.md does not specify what user data would be sent to a commercial API if enabled.
Install Mechanism: okNo install spec is provided (instruction-only + data + a single Python script). Nothing is downloaded or written to disk by an installer; risk from the install mechanism is minimal.
Credentials: noteThe skill requires no env vars in its registry metadata. SKILL.md documents optional CAREER_API_URL and CAREER_API_KEY for commercial mode — those env vars would be proportional to the advertised commercial feature. However, the metadata omission (they are not declared as required.env) and absence of code that uses them are inconsistent and should be resolved before providing keys to this skill.
Persistence & Privilege: okThe skill does not request persistent/always-on inclusion (always:false) and does not modify other skills or system settings. It does not request elevated privileges.