Skillv1.0.0

ClawScan security

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

Scanner verdict

SuspiciousFeb 14, 2026, 1:38 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's instructions and required tools are mostly consistent with its stated game-playing purpose, but important operational details are missing or unclear (notably authentication/authorization around joins, moves, and withdrawals), which raises risks when real crypto balances are involved.
Guidance: This skill is coherent in that it simply documents curl calls to a game API, but there are several things to check before using it with real funds: 1) Verify the operator and site (wirx.xyz) independently — check reputation, links to smart contracts, and on-chain token receipts. 2) Confirm how the backend authenticates and authorizes actions: the docs show only an agent name and registered wallet; ask whether requests require signed messages, API tokens, or other auth. If not, the system may be trivially spoofable. 3) Never share private keys or seed phrases. The skill does not request them, but some services later ask for signatures or keys — prefer on-chain signature flows via your wallet rather than pasting secrets. 4) Test with a throwaway/empty wallet and very small amounts first to confirm withdraws and swaps behave as documented. 5) Look for contract addresses and verify transfers on-chain (Solana/TON explorers) rather than trusting in-game balances alone. Because this is an instruction-only skill and makes network calls to an external site handling crypto, exercise caution and verify the service before using significant funds.

Review Dimensions

Purpose & Capability: noteThe name/description match the documented curl endpoints: joining, getting state, moving, checking balance, swapping, and withdrawing on wirx.xyz. Requiring only curl and no credentials is plausible for a central-server game that credits on-chain balances from a hot wallet. However, it's unusual that all actions in the examples are performed using only an agent name (and a wallet address at registration) with no authentication tokens or signatures — that omission is surprising for a crypto-related service and reduces confidence that the declared requirements capture what real usage will need.
Instruction Scope: concernSKILL.md is instruction-only and tells the agent to make curl calls to https://wirx.xyz endpoints. It does not instruct the agent to read local files or environment variables (good), but it also omits any authentication or signing steps for actions that affect balances (join/move/withdraw/swap). The API examples use only 'name' and sometimes 'wallet' — this suggests either the game trusts unauthenticated requests or the documentation is incomplete. That ambiguity could enable account spoofing, replay, or fund-misdirection attacks if the backend lacks proper auth.
Install Mechanism: okNo install spec and no code files — instruction-only skill that requires curl. This is low-risk from a code-install perspective because nothing is written to disk by the skill itself.
Credentials: okThe skill requests no environment variables, secrets, or config paths. That is proportionate given the presented API usage (which only shows wallet addresses). Because it does not ask for private keys or tokens, it avoids direct credential exfiltration risk — but the absence of any auth instructions is itself notable.
Persistence & Privilege: okThe skill does not request always:true and uses default invocation settings. It does not modify other skills or system settings. Autonomous invocation is allowed (default) but not a unique concern here.