github
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the official
ghcommand-line tool to interact with GitHub services. These operations are restricted to the permissions of the user's authenticated session. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to the processing of user-generated content from GitHub.
- Ingestion points: Untrusted data enters the agent context through the
gh pr view,gh issue list, andgh run viewcommands defined inSKILL.md. - Boundary markers: The instructions do not define delimiters or specific "ignore instructions" warnings when processing the fetched content.
- Capability inventory: The skill includes commands that can modify repository state, such as
gh pr create,gh pr merge, andgh issue create/close. - Sanitization: There is no explicit logic provided to sanitize or validate the content retrieved from GitHub before the agent processes it.
Audit Metadata