release
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes external data from task files and documentation to generate release notes.
- Ingestion points:
TASKS.mdand individual task files. - Boundary markers: Absent.
- Capability inventory: File system writes, Git operations (commit/tag), and repository hosting API interactions.
- Sanitization: None specified for task content.
- [COMMAND_EXECUTION]: The workflow involves executing git commands for version tagging and committing changelog updates, which are standard operations for release automation.
Audit Metadata