agent-architecture-designer

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the agent to utilize the Bash tool for validation tasks, including testing APIs with curl, building containers, and executing database performance analysis (EXPLAIN ANALYZE).
  • [EXTERNAL_DOWNLOADS]: The agent specification includes WebFetch and WebSearch tools, enabling the retrieval of content from external domains during the research phase.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes architectural requirements from an upstream agent (feature-analyst) while possessing high-privilege capabilities. * Ingestion points: Requirements and design specifications provided by the feature-analyst component in SKILL.md. * Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present. * Capability inventory: The agent has access to Bash, Read, Write, Edit, WebFetch, WebSearch, and Grep. * Sanitization: No evidence of input validation or sanitization for the data processed from the upstream source.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 01:13 AM