github-repo-setup
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh) andgitto automate repository management tasks, including creation, permission settings, and security feature enablement. These actions are transparent and appropriate for the skill's described purpose. - [EXTERNAL_DOWNLOADS]: Fetches
.gitignoretemplates fromhttps://www.toptal.com/developers/gitignore/api/. This is a well-known and reputable service for developers, and the download is restricted to static configuration files. - [PROMPT_INJECTION]: The skill incorporates user-provided inputs such as repository names, descriptions, and technology stacks into shell commands, creating a potential surface for indirect prompt injection.
- Ingestion points: User-supplied values for repository metadata in
SKILL.md. - Boundary markers: No explicit delimiters or instructions are used to isolate user input in the command templates.
- Capability inventory: Execution of shell commands (
gh,git,curl) and local file system writes. - Sanitization: None explicitly provided in the skill instructions; the system relies on the agent platform's internal input validation and safety controls.
Audit Metadata