gh
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the use of the official GitHub CLI (
gh) to perform various development and administrative actions, such as managing repositories and pull requests. These commands are executed via the shell. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of content from external sources that may be controlled by third parties, posing a risk of indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context through commands in
SKILL.mdthat retrieve descriptions, bodies, and comments from GitHub (e.g.,gh pr view,gh issue view,gh repo view). - Boundary markers: Absent. The instructions in
SKILL.mddo not include delimiters or specific instructions for the agent to treat the retrieved GitHub content as potentially untrusted. - Capability inventory: In
SKILL.md, the skill defines commands with significant side effects, including merging pull requests (gh pr merge), deleting repositories (gh repo delete), and making arbitrary API calls (gh api). - Sanitization: Absent. There is no mechanism described for sanitizing or filtering the content fetched from GitHub before it is processed by the agent.
Audit Metadata