Back to skill
Skillv1.0.0
ClawScan security
Lyrics Search · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 18, 2026, 2:50 PM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requested actions (a simple web fetch of a public lyrics API and minor text processing) match its description; it asks for no credentials, installs, or unusual privileges.
- Guidance
- This skill is internally consistent, but consider these practical points before installing: - Privacy: each lookup sends the song title/artist to a third-party public API (api.lrc.cx). If you need to keep queries private, avoid using it. - Copyright: displaying or printing full song lyrics may have copyright implications—verify you have the right to reproduce or distribute lyrics for your use case. - Reliability and rate limits: public APIs can be slow, rate-limited, or go offline; implement sensible timeouts and retries (as the SKILL.md suggests) and consider caching frequent queries. - Trustworthiness of API: the SKILL.md references api.lrc.cx but provides no homepage or publisher info — if provenance matters, validate the API service before heavy usage. - Autonomous calls: if you deploy this in an agent that operates autonomously, monitor usage to avoid unexpected repeated requests. If you need stronger guarantees, consider using a well-documented/official lyrics provider or hosting a vetted proxy/cache.
Review Dimensions
- Purpose & Capability
- okThe name/description (search song lyrics via LrcApi) aligns with the instructions: only a single GET to https://api.lrc.cx with title/artist parameters and simple text processing. No unrelated credentials, binaries, or paths are requested.
- Instruction Scope
- okSKILL.md only instructs the agent to perform a web_fetch to the LrcApi endpoint, optionally retry with artist omitted, strip timestamps/metadata, and format results for display/printing. It does not instruct reading system files, other env vars, or contacting other endpoints.
- Install Mechanism
- okThere is no install spec and no code files (instruction-only), which is the lowest-risk model — nothing is written to disk or installed by the skill itself.
- Credentials
- okThe skill declares no required environment variables, credentials, or config paths. Its functionality does not require elevated access or secrets, so the requested scope is proportionate.
- Persistence & Privilege
- okalways:false and default autonomous invocation are set. Autonomous invocation is the platform default and is reasonable for this type of skill; the skill does not request permanent presence or modify other skills/config.