set-up-forked-open-source-repo
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides standard developer utility for managing Git repository configurations. It does not exhibit malicious patterns or unexpected behaviors outside its stated purpose.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a coordinator script located at
${CLAUDE_PLUGIN_ROOT}/scripts/larch.shto perform Git and GitHub CLI operations. This execution is scoped to repository management and includes multiple anti-patterns (such as refusing to mutate remotes when classification is ambiguous) to ensure safe execution. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes external repository metadata and Git configuration strings.
- Ingestion points: Reads Git remote URLs (
origin.url) and repository metadata through thegh(GitHub) CLI. - Boundary markers: The skill documentation explicitly mentions a
phase_preflightcheck and the use ofnormalize_github_urlto validate inputs before fetching. - Capability inventory: The skill has the ability to write to local Git configurations and perform network operations via Git and the GitHub CLI.
- Sanitization: Inputs are sanitized via URL normalization and pre-flight classification to catch non-parseable or mixed-host remote layouts.
Audit Metadata