cpo-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze business context and product metrics (such as retention, engagement, and revenue) to generate strategic advice. This creates a surface where instructions embedded in external metrics data or business documentation could potentially influence the agent's strategic recommendations.
- Ingestion points: The skill reads business environment data from
company-context.mdand accepts product performance metrics via JSON input for the analysis scripts. - Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when interpolating or processing this external data.
- Capability inventory: The skill possesses the capability to read local files and execute bundled Python scripts (
pmf_scorer.py,portfolio_analyzer.py) to perform metric calculations. - Sanitization: The provided scripts use standard JSON parsing (
json.load()), which prevents code injection into the scripts themselves but does not sanitize the semantic content of the business data used for prompt generation.
Audit Metadata