dependabot-review
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection: The skill analyzes external content from pull request descriptions and package registry metadata to determine the impact of version bumps. This ingestion of untrusted data introduces a surface where strategically crafted text in a PR description or registry could influence the agent's analysis. * Ingestion points: The skill processes data from args.prBody and external metadata via the get_npm_package_info tool. * Boundary markers: There are no explicit instructions or delimiters defined to separate external data from the agent's internal logic. * Capability inventory: The agent uses tools like search_repo and read_repo_file to understand how dependencies are used in the codebase. * Sanitization: Content retrieved from pull requests and registries is processed without specific filtering or sanitization instructions.
Audit Metadata