production-parity-skill-builder
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from arbitrary repository files to build a parity map and generate new instructions, which could lead to the adoption of malicious directives found in code or documentation.\n
- Ingestion points: The skill reads
READMEfiles,docs/directories, application source code, CI/CD configurations (GitHub Actions, GitLab CI), and infrastructure manifests (Terraform, Kubernetes).\n - Boundary markers: There are no instructions for the agent to use delimiters or specific 'ignore' markers for content embedded within the files it inspects.\n
- Capability inventory: The skill is capable of reading repository files, executing local searches via
rg, and writing a newSKILL.mdfile to the project directory.\n - Sanitization: No sanitization or escaping protocols are defined for the repository data before it is interpolated into the generated skill's contract or surfaces.\n- [COMMAND_EXECUTION]: The skill utilizes
ripgrep(rg) to perform targeted discovery across the repository, searching for sensitive environment configurations, auth tokens, and infrastructure definitions.
Audit Metadata