public-roadmap-strategy
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided roadmap data, competitor benchmarks, and feedback board requests to generate strategy documents.\n
- Ingestion points: The skill ingests data during the 'Interview' phase (Step 1-8) and 'Workflow' steps in
SKILL.mdto define roadmap posture and vocabulary.\n - Boundary markers: No instructions are provided to use delimiters or sanitization techniques to distinguish between system instructions and untrusted roadmap content.\n
- Capability inventory: The skill advises writing to
developer-platform-context.mdfor long-term memory; however, it has no network access or shell execution capabilities.\n - Sanitization: The instructions lack requirements for the agent to validate or filter user-supplied data before it is saved to memory or included in generated policies.
Audit Metadata