design-agentic-architecture

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a standard instructional framework for software architecture and does not exhibit malicious behavior, obfuscation, or unauthorized data access.
  • [EXTERNAL_DOWNLOADS]: Step 7 requires the agent to perform web searches for current AI tools. This is a functional requirement to ensure the architecture design document contains up-to-date information, rather than a security risk.
  • [DATA_EXPOSURE]: The skill includes an instruction to save a generated design document to a local file path (Step 8). The default path provided is benign (./agentic-architecture-design.md), and the activity is consistent with the skill's purpose of document generation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external web searches (Step 7). While this is a potential ingestion point for untrusted data, the skill includes a 'false-confidence check' and specific auditing steps to verify the design against known traps, which serves as a safety mitigation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 07:37 PM
Security Audit — agent-trust-hub — design-agentic-architecture