pull-request
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
gh(GitHub CLI) command to push code and submit pull requests. This is consistent with the skill's primary purpose and utilizes a well-known developer tool. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it ingests untrusted data from code diffs to generate pull request metadata. 1. Ingestion points: Code modifications relative to the main branch are read to generate titles and descriptions in step n2. 2. Boundary markers: The instructions do not specify delimiters to separate the untrusted code content from the prompt context. 3. Capability inventory: The agent utilizes the
ghcommand-line tool to perform actions on GitHub, which involves network operations and repository modification. 4. Sanitization: There is no instruction to sanitize or filter the content of the diffs before they are processed by the agent.
Audit Metadata