architecture-analyzer

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation in VIOLATIONS.md includes shell commands using grep to search for architectural violations in the codebase, such as searching for specific class instantiations or database access patterns.
  • [PROMPT_INJECTION]: The skill performs analysis on project source code, which acts as a vector for indirect prompt injection where instructions embedded in the analyzed files could influence agent behavior.
  • Ingestion points: Local project files, particularly within the lib/ directory, are read via mcp__dart__analyze_files and grep.
  • Boundary markers: No specific delimiters or boundary markers are defined to isolate untrusted code content from the analysis instructions.
  • Capability inventory: The skill has access to file analysis tools (mcp__dart__analyze_files) and shell command execution (grep).
  • Sanitization: There is no evidence of sanitization or validation of the ingested code content before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 05:02 AM
Security Audit — agent-trust-hub — architecture-analyzer