argocd

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the official Argo CD CLI binary from the 'argoproj' organization's GitHub repository. The source is the authoritative repository for the tool described in the skill.
  • [COMMAND_EXECUTION]: Instructs the user to use sudo and chmod +x during the installation process in resources/install-and-setup.md. These commands are standard for installing system-wide CLI binaries on Linux and macOS.
  • [CREDENTIALS_UNSAFE]: Accesses the argocd-initial-admin-secret Kubernetes secret to retrieve the default administrative password. This is a required step for the initial setup and configuration of an Argo CD instance.
  • [PROMPT_INJECTION]: Presents a surface for indirect prompt injection as the skill is designed to process data from external Git repositories.
  • Ingestion points: External Git repositories specified in Application CRDs (e.g., resources/application-management.md).
  • Boundary markers: None explicitly defined in the prompts to differentiate between user instructions and data within manifests.
  • Capability inventory: Subprocess execution for argocd and kubectl commands (e.g., resources/command-cookbook.md).
  • Sanitization: Relies on standard Argo CD validation and CLI parsing; no additional agent-side sanitization is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 10:49 AM
Security Audit — agent-trust-hub — argocd