skills/kensaurus/skills/audit-cicd/Gen Agent Trust Hub

audit-cicd

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the gh CLI to perform administrative and diagnostic tasks via the GitHub API. It includes commands to delete non-expired artifacts and caches, which are destructive operations clearly marked as requiring user confirmation before execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from the GitHub API, such as workflow names and execution history, to generate its audit report. This ingestion of external content is a standard requirement for the skill's diagnostic functionality. Mandatory evidence: 1. Ingestion points: Output from gh repo list, gh api .../workflows, and gh api .../artifacts. 2. Boundary markers: None specified for the API output processing. 3. Capability inventory: Use of gh CLI for data retrieval and resource deletion. 4. Sanitization: No explicit sanitization of API responses before analysis.
  • [SAFE]: The skill encourages security best practices, including the identification and rotation of leaked secrets and the implementation of concurrency controls to minimize resource waste. All network interactions are conducted through the official GitHub CLI tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 11:26 PM