list-open-work
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes standard, read-only commands using
git,glab(GitLab CLI), andgh(GitHub CLI) to retrieve project metadata and pull request details. These operations are restricted to the context of the current repository and the user's own authored work. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as pull request titles and branch names to extract ticket identifiers (e.g., PROJ-123). This represents a vulnerability surface for indirect prompt injection; however, the skill applies strict regex-based extraction (
\b[A-Z][A-Z0-9]+-\d+\b) and limits the use of this data to generating static links and status text, which mitigates risk. - Ingestion points: PR titles, branch names, and descriptions retrieved via Git provider APIs.
- Boundary markers: None explicitly defined in the instructions.
- Capability inventory: Limited to read-only API calls and local Git metadata retrieval.
- Sanitization: Uses specific regex patterns for data extraction.
Audit Metadata