c4-model

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes 'npx likec4 validate' to ensure the generated architectural model is syntactically correct.
  • [EXTERNAL_DOWNLOADS]: Pulls the 'likec4' utility from the npm registry using npx during the validation phase.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by reading external architectural documentation and discovery briefs.
  • Ingestion points: Reads project files including 'docs/architecture/discovery-brief.md', ADRs, and LikeC4 model files.
  • Boundary markers: Lacks explicit delimiters or instructions to ignore potential commands embedded within architectural documentation.
  • Capability inventory: Performs file system operations (read/write) and executes shell commands via npx.
  • Sanitization: Relies on mandatory human approval of model diffs (Phase 5) before changes are written to the workspace.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 02:51 PM
Security Audit — agent-trust-hub — c4-model