skills/wenerme/ai/argocd-docs/Gen Agent Trust Hub

argocd-docs

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation provides standard instructions for installing Argo CD by downloading manifests and binaries from official project repositories (e.g., github.com/argoproj, quay.io). These sources are recognized as trusted vendor infrastructure.
  • [COMMAND_EXECUTION]: Includes a wide range of administrative command examples for kubectl, the argocd CLI, and basic shell operations. These are presented for educational purposes and use placeholders for environment-specific data.
  • [CREDENTIALS_UNSAFE]: Examples of Kubernetes Secrets and configuration files use dummy credentials such as EXAMPLE_PASSWORD, EXAMPLE_PRIVATE_KEY_PEM, and EXAMPLE_JWT. These are illustrative and do not expose real sensitive data.
  • [PROMPT_INJECTION]: The SKILL.md file includes instructions to the agent to verify manifests and CLI output against specific releases before providing answers. These are functional guidelines intended to ensure technical accuracy.
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests a large amount of external documentation, the primary ingestion point is controlled through SKILL.md instructions that treat all manifests and credentials as examples, and the lack of automated execution capabilities within the documentation itself minimizes the risk.
  • [SAFE]: The skill is authored by a known community contributor and mirrors official documentation. No malicious patterns, obfuscation, or persistence mechanisms were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 05:06 PM
Security Audit — agent-trust-hub — argocd-docs