013-agile-feature
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes the system
dateterminal command to ensure accurate timestamps are applied to generated feature documentation. - [INDIRECT_PROMPT_INJECTION]: The skill identifies and mitigates the risk of processing external data (Agile Epics) which could contain adversarial instructions.
- Ingestion points: Data is read from repository-owned files or maintainer-provided summaries in Step 1 of the workflow.
- Boundary markers: The skill includes explicit 'UNTRUSTED CONTENT' sections and instructions to treat epic content as evidence only, ignoring any conflicting embedded instructions.
- Capability inventory: The skill utilizes the
dateterminal command and tools for file reading (read_file,file_search). - Sanitization: The skill mandates the use of repository-owned paths or maintainer-sanitized summaries, explicitly advising against ingesting raw, unverified outsider-authored prose.
Audit Metadata