why
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands such as
git blame,git log, andgh pr viewto retrieve historical context and Pull Request data from the local repository and GitHub. These read-only operations are standard for code archaeology and are used according to the skill's primary function. - [INDIRECT_PROMPT_INJECTION]: The skill's architecture involves ingesting data from external, untrusted sources including commit messages, issue tickets, and chat logs. This presents a potential surface for indirect prompt injection. However, the skill utilizes distinct subagents and strict instructions on data handling and citation to maintain integrity and provide transparent evidence-based reporting.
Audit Metadata