gsdl-open-pr
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
gitand the GitHub CLI (gh) to perform repository operations such as pushing branches, creating pull requests, and committing code changes. These operations are intended for the skill's primary function. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting and acting on external content from GitHub pull request comments. * Ingestion points: External data enters the agent context via
gh pr viewandgh apicalls that fetch PR comments and review findings in SKILL.md. * Boundary markers: The skill provides instructions to triage findings against the project's SPEC.md to maintain scope, but lacks technical boundaries or escaping for the external comment data. * Capability inventory: The agent can modify the local filesystem, commit changes, push to remote repositories, and execute shell commands defined in the project's verification reports. * Sanitization: No programmatic sanitization or filtering is applied to the comment text before the agent evaluates it for potential code fixes.
Audit Metadata