ai-agents-architect
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- NO_CODE (SAFE): The skill consists entirely of markdown documentation. It includes conceptual code snippets for ReAct loops and tool registries but lacks any functional scripts (.py, .js, .sh), configuration files, or binary executables.
- PROMPT_INJECTION (SAFE): No instructions were found that attempt to bypass safety filters, override system instructions, or extract underlying prompts. The text is strictly instructional and follows its stated purpose as a systems architect guide.
- EXTERNAL_DOWNLOADS (SAFE): The skill does not reference external package repositories (npm, PyPI) or include commands to download and execute remote scripts.
- DATA_EXFILTRATION (SAFE): There are no patterns indicating access to sensitive local file paths or unauthorized network communication. The documentation does not handle user credentials or private data.
- INDIRECT_PROMPT_INJECTION (SAFE): As a static documentation skill, it lacks an ingestion surface for untrusted external data and does not possess write or execute capabilities that could be exploited via injection.
Audit Metadata