skills/peterbamuhigire/proposal-skills/proposal-storytelling-and-evaluator-journey/Gen Agent Trust Hub
proposal-storytelling-and-evaluator-journey
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external documents such as Terms of Reference (ToR) and evaluation criteria provided by third-party buyers, which represents an attack surface for indirect prompt injection.
- Ingestion points: External buyer documents, ToR, and project records defined in the 'Inputs' section of SKILL.md.
- Boundary markers: The workflow emphasizes inspecting evidence to 'record facts, assumptions, conflicts, and unavailable checks' and explicitly states 'stop on a load-bearing contradiction', providing a procedural boundary.
- Capability inventory: The 'Capability Contract' section limits the agent to read-only analysis and editing a requested working copy, with no permissions for network operations, system changes, or publishing.
- Sanitization: The workflow includes a 'critical-analysis and anti-slop gate' to verify all load-bearing claims against supplied evidence before release.
Audit Metadata