skills/kortix-ai/suna/kortix-rollback/Gen Agent Trust Hub

kortix-rollback

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill describes the use of administrative tools like gh and kubectl to perform production rollbacks. This includes triggering workflows in the kortix-ai/suna repository and patching Argo CD application configurations on a production EKS cluster (specifically arn:aws:eks:eu-west-2:935064898258:cluster/kortix-prod-eks). These high-privilege operations are appropriate for the skill's stated purpose of infrastructure management.- [PROMPT_INJECTION]: The skill documentation includes command templates that interpolate user-provided inputs directly into shell commands, creating a potential indirect prompt injection surface.
  • Ingestion points: The version and reason arguments used in the gh workflow run command template in SKILL.md.
  • Boundary markers: None are present around the interpolated values.
  • Capability inventory: Administrative shell access via gh and kubectl tools as documented in SKILL.md.
  • Sanitization: No sanitization or escaping mechanisms are specified for the user-supplied strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 11:48 AM
Security Audit — agent-trust-hub — kortix-rollback