task-planning
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for transforming external specifications into implementation tasks, creating a surface where malicious instructions in the source specification could influence generated output.
- Ingestion points: Ingests project specifications and implementation plans provided at runtime in SKILL.md.
- Boundary markers: No explicit delimiters are specified to isolate the untrusted specification data from the task generation instructions.
- Capability inventory: Mentions tools such as pytest, uv, mypy, and git for environment setup and verification in SKILL.md and modules/phase-structure.md.
- Sanitization: Input specifications are processed without explicit validation or escaping of their content.
- [SAFE]: The skill provides templates for .gitignore files in modules/tech-stack-patterns.md that include common sensitive file paths (e.g., .aws/, .env, *.pem) across multiple technology stacks. These are listed for the purpose of preventing accidental data exposure in version control systems and represent standard security best practices.
Audit Metadata