monorepo-ci-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a CI optimization strategy that processes untrusted data from the repository (package names and file paths).
- Ingestion points:
scripts/get-affected.tsreadspackage.jsonmetadata andgit diffoutput. - Boundary markers: Absent.
- Capability inventory: The script utilizes
child_process.execSyncfor shell operations, and the associated GitHub Action workflow uses the script's output in dynamic shell commands (pnpm --filter ${{ matrix.package }} test). - Sanitization: Absent. There is no validation or escaping of package names retrieved from
package.jsonbefore they are interpolated into shell commands. - [COMMAND_EXECUTION]: The provided TypeScript script uses
child_process.execSyncto execute git commands. The skill also provides GitHub Action templates that construct shell commands dynamically based on file system metadata. - [EXTERNAL_DOWNLOADS]: The skill utilizes official GitHub Actions (e.g.,
actions/checkout,actions/setup-node) and integrates with well-known services like Nx Cloud and Turborepo. These references target trusted organizations and represent standard CI/CD practices.
Audit Metadata