idea-generation
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs legitimate financial analysis and research tasks using a multi-step validation funnel.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection via external data ingestion, which is deemed safe given its primary research purpose and the structured triage framework.
- Ingestion points: Reads news and catalysts via
WebSearchandWebFetchtools, and processes user-supplied portfolio/watchlist files. - Boundary markers: No explicit delimiters for external content are defined in the instructions, but Step 5 and Step 6 require the agent to validate findings against specific economic archetypes and peer metrics.
- Capability inventory: Accesses financial MCP tools (e.g.,
screen_stocks,get_financial_statements) and performs file operations in the{task}/directory. - Sanitization: While no technical sanitization is mentioned, the skill mandates rigorous cross-checking against sector-specific bases and historical metrics, reducing the risk of accidental obedience to malicious instructions.
Audit Metadata