suede-ci-gate
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local repository manifests to generate configuration files and CLI commands.
- Ingestion points: The skill reads package.json, pyproject.toml, go.mod, and other manifests in SKILL.md (Step 0) to detect project structure and runtime versions.
- Boundary markers: No delimiters or "ignore embedded instructions" warnings are used when processing the manifest data.
- Capability inventory: The skill creates .github/workflows/ files and generates gh api CLI commands.
- Sanitization: The instructions do not specify any validation or escaping of the values extracted from the manifests before interpolation.
- [EXTERNAL_DOWNLOADS]: Generated workflows reference external GitHub Actions such as dorny/paths-filter and actions/checkout to handle path filtering and repository access.
- [COMMAND_EXECUTION]: The skill generates gh api command templates for managing repository branch protection, which are intended for user execution.
Audit Metadata