architecture-analyzer
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill analyzes untrusted source code, creating a surface for indirect prompt injection via code comments designed to mislead the analyzer. \n
- Ingestion points: The agent reads project files using
mcp__dart__analyze_filesandgrepcommands. \n - Boundary markers: Absent. No explicit instructions are provided to distinguish between code logic and potential malicious instructions in comments. \n
- Capability inventory: Static analysis and local command execution (
grep). \n - Sanitization: Absent. Findings from code files are reflected directly in the generated analysis reports.\n- [Command Execution] (SAFE): The skill utilizes
grepand MCP tools for local static analysis within the project directory. This is consistent with the primary purpose of an architecture analyzer and does not involve elevated privileges or remote network access.
Audit Metadata