crypto-trading-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the AI agent to fetch and analyze information from third-party data sources such as Coinglass, DefiLlama, LunarCrush, and others. This creates a surface where malicious instructions could be embedded in the fetched content (e.g., social media sentiment or news) to influence the agent's trading advice or behavior.
- Ingestion points: External data sources listed in
SKILL.md, including market sentiment, derivatives data, and on-chain metrics. - Boundary markers: The skill lacks explicit instructions or delimiters to tell the agent to ignore any embedded commands within the external data it processes.
- Capability inventory: No dangerous scripts or system-level capabilities are included within the skill; it primarily focuses on data analysis and providing recommendations.
- Sanitization: There are no provided mechanisms to sanitize or filter the external data before it is presented to the agent for analysis.
- [EXTERNAL_DOWNLOADS]: The
bin/index.jsscript, executed during installation vianpx, downloads theSKILL.mdcontent from the author's public GitHub repository (github.com/0xrikt/crypto-skills). This is a standard installation procedure for this skill and targets the vendor's own infrastructure.
Audit Metadata