Skillv1.0.0

ClawScan security

Meme Token Detector · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 6, 2026, 12:27 AM

Verdict: suspicious
Confidence: high
Model: gpt-5-mini
Summary: The skill claims to perform on-chain rug-pull detection and charge 0.001 USDT, but the shipped code is a hardcoded stub and there is no implemented payment or blockchain-analysis integration—do not send funds and treat the skill with caution.
Guidance: This skill's docs say it charges 0.001 USDT and list a payment wallet, but the shipped Python script is only a hardcoded demo that never performs real blockchain checks or collects/validates payment. Do not send funds to the listed wallet. If you consider using this skill: (1) ask the author to show the real integration (which APIs/indexers are called and where payments are handled), (2) request source that actually queries on-chain data and/or a transparent server-side payment flow, (3) review the full code yourself or have it audited to confirm it performs the promised analysis, and (4) test locally with known contract addresses before trusting outputs. Because of the mismatch between claims and implementation, treat this skill as untrusted until the author provides a verifiable implementation of both analysis and payment handling.

Review Dimensions

Purpose & Capability: concernThe name/description promise comprehensive on-chain analysis (liquidity locks, minting/backdoors, tax, holder distribution, audit status) and an automated paid service. However, the included script contains only a hardcoded example result and does not call any blockchain APIs, indexers, or payment processors. The presence of a payment claim (wallet address and text '已扣费 0.001 USDT') is not implemented in code. This mismatch means the skill does not actually provide the claimed capabilities.
Instruction Scope: concernSKILL.md instructs users to run the local Python script and explicitly states a charge and provides a wallet address for payment. The runtime instructions do not implement or verify any payment, nor do they call any external chain services; they only print a simulated report. This creates risk that users will be asked (in docs) to send money for a service that the code does not deliver. The instructions do not attempt to access unrelated files or secrets, which is good, but the payment claim is misleading and potentially a scam vector.
Install Mechanism: okNo install spec; the skill is instruction-only with a small script. Nothing is downloaded or installed automatically, so there is low installation risk.
Credentials: okNo environment variables, credentials, or config paths are requested. The skill does include a hardcoded public wallet address in SKILL.md for payments, but it does not request tokens or secrets. The lack of required credentials is proportionate to the actual (minimal) implementation.
Persistence & Privilege: okalways is false and the skill requests no elevated privileges or persistent presence. It does not modify other skills or system settings. Autonomous invocation is allowed by platform defaults but does not combine with other red flags here.