why
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local CLI tools including
git(blame, log, show) andgh(GitHub CLI) to extract historical context and pull request details from the development environment. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it ingests and processes data from multiple untrusted external sources.
- Ingestion points: Processes data from PR descriptions and comments, task descriptions and status updates (Linear/Jira), internal long-form documentation (Notion/Confluence), real-time team conversations (Slack/Teams), and incident postmortems.
- Boundary markers: Absent; the instructions for subagents rely on stylistic
Audit Metadata