milestone-review
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from various project files and codebase comments to generate milestone reviews, creating a potential vector for indirect instructions.
- Ingestion points: Data is read from files in
production/milestones/,production/sprints/,production/risk-register/, and codebase markers (TODO, FIXME, HACK) via search tools. - Boundary markers: The instructions do not define boundary markers or explicit delimiters to separate untrusted content from the agent's instructions.
- Capability inventory: The skill utilizes
Read,Glob,Grep,Write,Task(for sub-task delegation), andAskUserQuestiontools. - Sanitization: There is no evidence of sanitization, validation, or escaping of the content ingested from the codebase or project reports before it is processed by the agent.
Audit Metadata