operational-readiness
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [PROMPT_INJECTION] (SAFE): The instructions are strictly task-oriented and do not contain attempts to bypass safety filters, override system prompts, or extract system-level instructions.
- [DATA_EXFILTRATION] (SAFE): Although the skill accesses sensitive files (CI/CD configs, Terraform, secrets management paths) to perform its audit, it lacks network capabilities (like curl or fetch) to transmit this data externally. All analysis remains within the user's session.
- [COMMAND_EXECUTION] (SAFE): The workflow utilizes read-only utilities such as Grep and Glob for pattern matching. There are no instructions to execute arbitrary shell commands or modify system files.
- [REMOTE_CODE_EXECUTION] (SAFE): The skill does not download external scripts, install packages, or perform runtime compilation.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill ingests untrusted data from the codebase (e.g., scanning .tf and .yml files). While a malicious file could theoretically attempt to confuse the auditor with misleading patterns, the skill has no 'write' or 'execute' capabilities that could be exploited through such an injection. The only output is a markdown report.
- [CREDENTIALS_UNSAFE] (SAFE): The skill specifically audits for hardcoded secrets and identifies the presence of secrets managers, which is a defensive security feature. It does not contain any hardcoded credentials itself.
Audit Metadata