Back to skill
Skillv1.0.0
ClawScan security
gate-exchange-welfare · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 6, 2026, 11:00 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's declared requirements and runtime instructions are coherent with a read-only Gate welfare lookup: it only needs Gate API credentials to call identity and beginner-task read endpoints and contains no hidden installers or unrelated secrets.
- Guidance
- This skill appears to do exactly what it says: read-only queries for Gate welfare info using your Gate API key/secret. Before installing, ensure you: (1) only provide credentials via your configured MCP/session (do not paste secrets into chat), (2) grant the minimum permission (Welfare:Read) and rotate keys if you later remove the skill, and (3) confirm the skill repository/source is trustworthy (the bundle references github.com/gate/gate-skills). If you need stronger assurance, ask the publisher for an official audit or verify the GitHub source and commit history before use.
Review Dimensions
- Purpose & Capability
- okName/description, declared MCP tools (identity + beginner task list), and required env vars (GATE_API_KEY, GATE_API_SECRET) align with a welfare-center read-only query skill — the requested credentials are appropriate for the stated API calls.
- Instruction Scope
- okSKILL.md restricts operations to two read-only MCP tools, enforces runtime rules (do not fabricate data, do not call undocumented tools), and explicitly forbids asking users to paste secrets into chat. It does not instruct reading unrelated files or transmitting data to unexpected endpoints.
- Install Mechanism
- okInstruction-only skill with no install spec and no code files present; nothing is downloaded or written to disk during installation, minimizing install-time risk.
- Credentials
- okOnly two environment variables (GATE_API_KEY, GATE_API_SECRET) are required, matching the primary credential and the read-only Welfare:Read permission. No unrelated credentials, config paths, or broad secrets are requested.
- Persistence & Privilege
- okSkill is not always-enabled and is user-invocable; it does not request elevated persistent privileges. Autonomous invocation is allowed (platform default) but the skill's scope is read-only and limited.