github-to-origin
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform extensive repository and deployment reconfigurations through shell commands using
git,vercel, andoriginCLI tools. These operations include modifying remote URLs, changing project linkings, and disconnecting/reconnecting git providers.- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input which is then used to construct shell commands, creating a vulnerability surface. - Ingestion points: User-provided repository identifiers (org/repo) and production domain names collected at the start of the runbook in
SKILL.md. - Boundary markers: Absent; the instructions do not define delimiters or warnings to treat these specific inputs as literal data only.
- Capability inventory: High-privilege CLI operations across
git,vercel, andorigin, including repository creation and deployment pipeline modification. - Sanitization: Absent; there are no instructions for the agent to validate, escape, or sanitize the user-provided strings before they are executed within the shell environment.
Audit Metadata