Skillv1.2.0

ClawScan security

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

BenignFeb 15, 2026, 4:12 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's declared purpose (interacting with the Signet trust API) matches its requirements and instructions: it only needs a SIGNET_API_KEY and provides HTTP endpoints for the described operations.
Guidance: This skill appears coherent and only needs a Signet API key to call the documented endpoints. Before installing: (1) Confirm you trust the Signet service and the skill author (registry source is unknown here). (2) Treat SIGNET_API_KEY like any secret: only set it for skills you trust and rotate it if you suspect misuse. (3) Because the skill is instruction-only, the runtime agent that uses this skill will have access to the SIGNET_API_KEY when invoked — ensure you trust the agent and its ability to call external APIs. (4) If you need stronger assurance, request a skill from a verified publisher or inspect an implementation (code) so you can see exactly how the key is used.

Purpose & Capability: okName/description describe a Signet trust-id API client and the skill only requests a SIGNET_API_KEY and documents Signet HTTP endpoints — these requirements are proportional and expected for this purpose.
Instruction Scope: okSKILL.md is an instruction-only API spec showing unauthenticated registration endpoints and authenticated endpoints that use Authorization: Bearer $SIGNET_API_KEY. The instructions do not ask the agent to read unrelated files or other environment variables, nor to transmit data to unknown endpoints; they explicitly instruct storing the api_key in SIGNET_API_KEY and not sending it to services other than api.agentsignet.com.
Install Mechanism: okNo install spec and no code files are present (instruction-only). This minimizes on-disk execution risk and is appropriate for a pure API-integration skill.
Credentials: okOnly a single credential (SIGNET_API_KEY) is required and it directly maps to the documented authenticated endpoints. The number and type of env vars requested are proportionate to the skill's stated function.
Persistence & Privilege: okThe skill is not force-included (always: false), does not request system config paths, and has no install steps. It does not ask to modify other skills or global agent settings.