skills/sunic4/dev-skills/cc-feat/Gen Agent Trust Hub

cc-feat

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill incorporates a comprehensive internal security audit phase (S1-S5) triggered by sensitive keywords and behaviors, ensuring that generated code is reviewed for authentication, input validation, and data handling issues.\n- [COMMAND_EXECUTION]: Local utilities such as grep, read-yaml, and validate-yaml are used to manage project state and verify compliance with architectural roadmaps.\n- [PROMPT_INJECTION]: The skill processes untrusted input from external requirement and architecture documents, which constitutes an indirect prompt injection surface.\n
  • Ingestion points: Requirements (REQ), architecture (ARCH) documents, and items.yaml files are read into the agent's context during the design and implementation phases.\n
  • Boundary markers: There are no explicit instructions or delimiters defined to isolate the content of these external documents from the skill's operational instructions.\n
  • Capability inventory: The skill has the ability to write source code and execute local scripts like review-generate.mjs.\n
  • Sanitization: No explicit content filtering is performed on ingested technical documents prior to their use in code generation, although the output code is subject to a secondary internal audit.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 06:03 AM
Security Audit — agent-trust-hub — cc-feat