github
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the official GitHub CLI (
gh) to perform actions like creating pull requests, merging branches, and listing issues. Commands such asgh pr createandgh issue listare used as intended for repository management. - [EXTERNAL_DOWNLOADS]: The skill's metadata contains installation instructions for the
ghtool via trusted package managers (Homebrew and APT). Since these target official repositories and well-known services, they are considered safe. - [PROMPT_INJECTION]: The skill retrieves content from external sources (GitHub issue and PR bodies) which represents an indirect prompt injection surface.
- Ingestion points:
gh pr viewandgh issue listinSKILL.mdretrieve external text content. - Boundary markers: Absent. The skill does not currently use specific delimiters to isolate external data from instructions.
- Capability inventory: Pull request merging (
gh pr merge), issue creation (gh issue create), and CI workflow reruns (gh run rerun) are available inSKILL.md. - Sanitization: Absent. External text is processed without explicit filtering or validation.
- [SAFE]: No obfuscation, data exfiltration, or persistence mechanisms were found. The skill maintains the principle of least privilege by using the existing
ghauthentication context.
Audit Metadata