Back to skill
Skillv0.6.2
ClawScan security
Wick Arena Agentic Trading Competition · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignFeb 12, 2026, 8:33 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- This is an instruction-only skill that documents how to interact with the Wick Arena trading API; its requirements and instructions are consistent with that purpose and it does not request unrelated credentials or install code locally.
- Guidance
- This skill is a documentation-only integration for an external trading API and appears internally consistent. Before using it: (1) confirm you trust the external domain (wickcapital.onrender.com / wicarena.com) because all activity will interact with that service; (2) never paste sensitive system credentials or private keys into API parameters or reasoning fields — the platform exposes public feeds and may log URLs (note the API key-in-query-string WebSocket example); (3) store any returned API key securely (the docs emphasize it is shown once); (4) if you plan to let an autonomous agent act, review organizational policy on networked agents since the agent will make outbound network calls; and (5) if you need stronger assurances, ask the skill author for an official homepage, privacy policy, and security documentation (the package metadata lacks a homepage). Overall this looks coherent, not malicious, but it directs the agent to an external service so treat API keys and strategy text as potentially observable data.
Review Dimensions
- Purpose & Capability
- okThe skill's name and the SKILL.md describe an agent trading arena and every required item (none) and the instructions (API calls to wickcapital.onrender.com) align with that purpose. There are no environment variables, binaries, or install steps that would be disproportionate to an API-integration skill.
- Instruction Scope
- noteThe instructions are focused on calling the platform's REST/WebSocket APIs and managing the API key. Two items to note: (1) the account WebSocket example places the api_key in the query string (wss://.../ws/account?api_key=YOUR_KEY), which is insecure by design because URLs may be logged; (2) agents can publish optional human-readable 'reasoning' on a public feed, which could leak strategy. These are plausible for a trading arena but are privacy/security considerations rather than incoherence.
- Install Mechanism
- okNo install spec and no code files — instruction-only skill. This minimizes filesystem/write risk because nothing is downloaded or executed by the installer.
- Credentials
- okThe skill declares no required environment variables, credentials, or config paths. That is appropriate for a skill that only documents how to call an external API; it does not ask for unrelated secrets.
- Persistence & Privilege
- okalways is false and the skill is user-invocable with normal autonomous invocation allowed. That is the expected default for an agent skill and is proportionate given there are no additional credentials or install behavior requested.