review-dependabot
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) and the Bun runtime (bun) to perform repository operations, check pull request statuses, and run local builds or tests. These operations are standard for development workflows but represent the primary capability surface of the skill.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and interpret pull request bodies, which are external, untrusted data sources. • Ingestion points: Pull request bodies are retrieved using 'gh pr view' and used as the primary source for changelog triage and decision-making in SKILL.md. • Boundary markers: Absent. There are no instructions to the agent to use delimiters or to treat the ingested text as potentially containing adversarial instructions. • Capability inventory: The agent has the ability to execute terminal commands (gh, bun) and perform file system operations during the review process. • Sanitization: Absent. The skill does not specify any validation, filtering, or escaping of the content retrieved from pull request bodies.
Audit Metadata