redmine-projects

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): The skill file consists solely of documentation and configuration for Redmine integration. No malicious instructions or suspicious code patterns were found.
  • [NO_CODE] (SAFE): This skill defines tool interfaces and workflows but does not include any Python, JavaScript, or shell scripts, significantly reducing the attack surface.
  • [Indirect Prompt Injection] (LOW):
  • Ingestion points: The tools redmine_list_projects, redmine_get_project, and redmine_get_project_versions ingest data from an external Redmine instance (project names, descriptions, and version details).
  • Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined in the workflow documentation.
  • Capability inventory: The skill documentation does not define any high-risk capabilities such as arbitrary command execution, file system writes, or external network requests beyond the intended Redmine API calls.
  • Sanitization: No explicit sanitization or validation logic is mentioned for the data retrieved from Redmine.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:33 PM