business-evaluation
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a standard business analysis workflow. It reads from local project paths (e.g., market analysis files) and writes reports to the
docs/directory using provided templates. - [PROMPT_INJECTION]: The skill uses strong instructional markers (e.g.,
MANDATORY IMPORTANT MUST ATTENTION,[BLOCKING],NEVER skip) to enforce reliability and prevent the model from skipping steps or using unverified data. These are intended for protocol compliance rather than bypassing system safety guardrails. - [COMMAND_EXECUTION]: The skill references the
spawn_agenttool and internal sub-skills (e.g.,$market-analysis,$domain-analysis). These are standard platform capabilities for task delegation and workflow management in the target environment. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from a
MARKET_ANALYSIS_PATHfile and user input. While this is an ingestion surface, the skill limits the impact by requiring cited evidence and providing a confidence score for its output, which serves as a manual review checkpoint. - [DATA_EXPOSURE]: No sensitive file paths (like SSH keys or AWS credentials) or hardcoded secrets were detected. The file access is restricted to project-specific documentation.
Audit Metadata