forex-list
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (LOW): The skill's setup instructions and workflow involve downloading and executing the
octagon-mcppackage from npm usingnpx. Since the 'OctagonAI' organization is not among the predefined trusted sources, this constitutes an unverifiable dependency. The severity is lowered to LOW as this is the primary intended purpose of the skill. - [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. The skill relies on the
octagon-agenttool to retrieve data from external sources (web searches and other agents), which could theoretically contain malicious instructions designed to influence the agent. - Ingestion points: Data returned by the
octagon-agenttool. - Boundary markers: Absent; no specific delimiters or 'ignore' instructions are provided for the data processing step.
- Capability inventory: The agent can execute MCP calls to the
octagon-mcpserver. - Sanitization: No evidence of sanitization or validation of the retrieved external content.
- [COMMAND_EXECUTION] (SAFE): The documentation (mcp-setup.md) provides standard installation commands for Homebrew and Node.js. These are common developer tools and the instructions are intended for manual user setup rather than automated execution by the skill.
Audit Metadata