skills/product-on-purpose/thinking-framework-skills/think-reference-class-forecasting/Gen Agent Trust Hub
think-reference-class-forecasting
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for the agent to analyze user-provided project data, which constitutes a data ingestion surface.
- Ingestion points: User-provided project details, duration, and success odds as described in SKILL.md.
- Boundary markers: Instructions explicitly mandate using real base-rate data and refuse to fabricate numbers.
- Capability inventory: The skill defines no tool usage, network operations, or file-writing capabilities across any scripts.
- Sanitization: No specific input sanitization is present, but the agent is directed to follow a strict output template.
- [SAFE]: Detailed inspection of all skill files, including instructions and metadata, found no malicious patterns, obfuscation, or remote code execution vulnerabilities. The skill is a purely informational reasoning guide.
Audit Metadata