pick-for-the-problem
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The instructions establish a cognitive protocol for selecting tools and libraries based on problem requirements rather than reflex. No malicious code, network exfiltration, or persistence mechanisms are present.
- [PROMPT_INJECTION]: The skill defines a surface for processing user-supplied technology scenarios. 1. Ingestion points: User prompts describing a technology choice or dependency (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: The skill specifies routing to internal platform research tools (/discovery:research) and reading internal context files. 4. Sanitization: Absent. Although explicit delimiters are missing, the skill's core purpose is to apply critical auditing and skepticism to user justifications, which serves as a natural defensive measure against manipulative inputs. (Severity: LOW).
Audit Metadata