token-movers
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple external sources, including cryptocurrency token names and social media sentiment from X.ai search results. This data is interpolated into markdown reports and logs, presenting a surface for indirect prompt injection if the external data contains malicious instructions.
- Ingestion points: External market data and social sentiment APIs accessed in SKILL.md.
- Boundary markers: The instructions lack explicit structural boundaries or delimiters for external content in the final report output.
- Capability inventory: Subprocess execution for data fetching, file system writes to output and memory directories, and network access to external domains.
- Sanitization: Relies on basic filtering instructions to exclude bots and shill posts from social media content.
- [EXTERNAL_DOWNLOADS]: The skill downloads cryptocurrency market data, token information, and social media sentiment from well-known services such as CoinGecko, GeckoTerminal, DexScreener, and X.ai. These downloads are central to the skill's documented purpose.
- [COMMAND_EXECUTION]: Automates the retrieval and processing of market data using shell commands, utilizing tools like curl, jq, and grep to parse JSON and filter information.
- [DATA_EXFILTRATION]: Communicates with external APIs to fetch token data. These network operations target reputable service providers and do not involve the transmission of sensitive local environment data or user credentials.
Audit Metadata