strategy-researcher
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user strategy ideas and market data to produce Strategy Hypothesis documents intended for consumption by subsequent AI agents (e.g., a coder agent). This creates a surface where malicious user input could attempt to influence the logic or behavior of the downstream agents in the pipeline.
- Ingestion points: SKILL.md dynamic execution zone (user queries and strategy idea exploration modes).
- Boundary markers: Present. The skill utilizes explicit Static Context Block markers and workflow phase delimiters to isolate instructions from session data.
- Capability inventory: None. The skill generates documentation and does not contain functionality for code execution, network requests, or file system access.
- Sanitization: The skill includes robust validation logic, including a Decision Tree and Validation Gates that require numerical thresholds and unambiguous rules to ensure output quality.
Audit Metadata