Skillv1.0.1

ClawScan security

Guiro · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 19, 2026, 5:42 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's requirements, scripts, and instructions align with its stated purpose (publishing A2UI JSON to guiro.io using an API key); nothing in the package requests unrelated credentials or performs unexpected actions.
Guidance: This skill appears coherent and focused on publishing A2UI bundles to guiro.io. Before installing: (1) Treat GUIRO_API_KEY as a secret and only provide a key scoped minimally for create/validate operations; (2) avoid sending sensitive PII or secrets in payloads because created snapshots are publicly viewable via short links; (3) note the scripts save a capabilities file under .guiro/runtime-capabilities.json and will POST payload.json to api.guiro.io; (4) if you need tighter control, review the API provider's retention and access policies and consider using a dedicated API key you can rotate if needed.

Purpose & Capability: okThe skill claims to publish A2UI bundles to guiro.io and requires curl, a shell, and GUIRO_API_KEY — all of which are appropriate and expected for making authenticated HTTP requests to the guiro API. No unrelated services or credentials are requested.
Instruction Scope: okSKILL.md instructs the agent to fetch capabilities, build/validate a JSON payload, and POST it to the API using the provided scripts. The scripts only read the payload file and an optional capabilities cache (.guiro/runtime-capabilities.json) and do not attempt to read other system files, credentials, or network endpoints beyond api.guiro.io.
Install Mechanism: okThere is no install spec (instruction-only with included scripts). No external downloads, packages, or extract operations are performed by the skill package itself. This minimizes disk writes to the shipped scripts and any files the scripts explicitly create (e.g., .guiro/runtime-capabilities.json).
Credentials: okOnly a single API credential (GUIRO_API_KEY) is required and is used consistently in the scripts. A few optional environment variables (GUIRO_PREFLIGHT_CAPABILITIES, GUIRO_CAPABILITIES_FILE) are present to control behavior; nothing else sensitive or unrelated is requested.
Persistence & Privilege: okThe skill is not always-enabled, does not modify other skills, and only writes a capabilities cache under .guiro. It does not request elevated privileges or persistent system-wide changes.