surf-ingress-migration

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute several shell commands involving cluster management and repository operations. These include kubectl for resource inspection, helm template for manifest generation, git for branch and commit management, and curl for connectivity testing. While these are necessary for the skill's stated purpose of ingress migration, they represent a significant capability tier that requires careful oversight.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the use of unvalidated user input (specifically the tenant/app name) within shell command templates.
  • Ingestion points: The tenant name provided by the user during invocation or through subsequent interaction.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded control characters are provided in the command templates.
  • Capability inventory: Includes the ability to perform file system writes (via git/manifest edits), network operations (via curl), and cluster configuration changes (via kubectl/helm).
  • Sanitization: There is no evidence of sanitization or escaping logic for the <app> or <env> placeholders before they are interpolated into shell commands like git checkout -b feat/kong-to-traefik-<app> or kubectl get svc -n services-<app>.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 09:51 PM