github-pr-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The main script
Invoke-GitHubPRWorkflow.ps1executes standard development tools includinggitandgh(GitHub CLI) to manage branches and pull requests. This behavior is consistent with the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: The skill includes functionality to install the GitHub CLI using well-known Windows package managers like
wingetandscoop. These are considered safe distribution channels for development tools. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface by processing data from
git logand user inputs to generate PR descriptions. - Ingestion points: User inputs and
git logoutput are processed inInvoke-GitHubPRWorkflow.ps1. - Boundary markers: No specific delimiters or warnings for the agent are used when interpolating this data into commands.
- Capability inventory: The skill can execute shell commands through the GitHub CLI and Git as defined in the primary script.
- Sanitization: The script relies on standard shell argument handling and does not perform additional sanitization on the ingested text content.
Audit Metadata