Skillv1.0.0

ClawScan security

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

Scanner verdict

SuspiciousMar 6, 2026, 12:28 AM

Verdict: suspicious
Confidence: high
Model: gpt-5-mini
Summary: The skill claims to decode contract calls and internal transfers and charge 0.001 USDT per call, but the included code only fetches basic transaction fields from Etherscan (no input/function decoding or internal transfer analysis) and there is no payment infra — review before use.
Guidance: This skill is suspicious because it advertises full transaction decoding and automatic charging but the code only fetches basic tx fields from Etherscan and does not implement payment or advanced decoding. Before installing: (1) review and understand the included script — it only performs a single HTTP GET to api.etherscan.io and truncates fields, it does not decode function signatures or internal transfers; (2) be aware that queries (transaction hashes) are sent to Etherscan and are visible to that service; (3) the stated per-call charge (0.001 USDT) is not enforced by the code — if you need billing, verify incoming payment handling separately; (4) ensure the runtime has Python and the 'requests' library or add an install step; (5) because the owner/homepage are unknown, prefer to audit/modify the code (e.g., add an API key parameter, proper decoding via ABI/4byte/database and internal tx lookups) or obtain a vetted alternative from a known source before use.

Review Dimensions

Purpose & Capability: concernThe skill's description and SKILL.md claim full transaction decoding (contract function names, input decoding, internal transfers, gas cost calculation). The included script only performs a single Etherscan proxy call to fetch basic transaction fields (hash, from, to, value, gas, input truncated) and does not decode input data, resolve function signatures, or analyze internal transactions. The claimed capabilities are therefore not implemented by the provided code.
Instruction Scope: concernSKILL.md states features (decode input, show called contract function, analyze internal transfers) but provides no runtime instructions for paying/charging users or using an external service; the script calls api.etherscan.io only and does not implement payment or deeper decoding. The runtime behavior (network request to Etherscan) is narrower than the advertised scope, so consumers may be misled about what the skill does.
Install Mechanism: noteThere is no install spec (instruction-only + small script). This is low-risk from an install perspective, but the script depends on the Python 'requests' package which is not declared; running the script may fail if requests is not available. No downloads or third-party installers are used.
Credentials: okThe skill requests no environment variables, credentials, or config paths — this is proportional to its actual behavior (a simple public HTTP call). Note: the script sends transaction hashes to Etherscan (third‑party endpoint), so queries are visible to that service.
Persistence & Privilege: okThe skill does not request always:true, does not modify other skills, and does not require persistent/system-wide privileges. Autonomous invocation is allowed (platform default) but not combined with other high-risk flags.