deploy

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the openshift MCP server's resources_create_or_update capability to modify cluster state. This high-privilege operation is secured by mandatory human-in-the-loop checkpoints that require the user to review and approve rendered YAML manifests for deployments, services, and routes before execution.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface in its port detection logic. 1. Ingestion points: It parses local project files including Dockerfile, nginx.conf, httpd.conf, and package.json. 2. Boundary markers: No explicit boundary markers are used during the file reading phase. 3. Capability inventory: The skill has the ability to create/update cluster resources and access pod logs via the openshift MCP. 4. Sanitization: All detected values are presented to the user for validation before being injected into deployment templates, and the skill uses structured YAML templates for resource generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 04:04 PM