project-planner
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input such as feature requests and requirements.
- Ingestion points: Requirements analysis phase in SKILL.md.
- Boundary markers: No specific delimiters or ignore instructions for input requirement data.
- Capability inventory: File system tools (Read, Grep, Glob, Write) for analysis and plan generation.
- Sanitization: Input is not sanitized, but the structured output template serves as a control mechanism.
- [SAFE]: The skill provides a detailed Risk Assessment Checklist which advises users on preventing security vulnerabilities like injection attacks and unauthorized data access in their implementations.
- [SAFE]: The skill uses limited tool permissions and contains no indicators of malicious behavior such as obfuscation or exfiltration.
Audit Metadata