subagent-driven-development
Pass
Audited by Gen Agent Trust Hub on May 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its ingestion of implementation plans and subagent reports.
- Ingestion points: External data enters the context via the plan files extracted in SKILL.md and the implementation reports processed in spec-reviewer-prompt.md.
- Boundary markers: The prompt templates utilize Markdown headers (e.g., '## Task Description', '## What Was Requested') as delimiters to separate instructions from untrusted content.
- Capability inventory: The subagents possess high-privilege capabilities including the ability to write source code, execute tests, and perform git commits via platform-provided tools.
- Sanitization: The skill does not implement explicit sanitization, validation, or escaping of the task descriptions or subagent reports before interpolating them into subsequent subagent prompts.
Audit Metadata