langgraph-adk

Pass

Audited by Gen Agent Trust Hub on Jun 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands via Bun to manage project lifecycles, including installing dependencies (bun add), verifying code integrity (bun run typecheck), and running the generated application (bun run start). these are standard operations for a developer tool.
  • [EXTERNAL_DOWNLOADS]: The skill initiates downloads of standard JavaScript/TypeScript packages from the NPM registry to provide the necessary runtime environment for the generated agents (e.g., langchain, zod, ink).
  • [SAFE]: The skill functions as a legitimate code generation and scaffolding tool. It uses environment-based configuration for LLM endpoints and retrieves model metadata (such as context window size) using standard API requests to the user-specified base URL. The workflow includes architectural design steps and diagnostic checks that align with its stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 9, 2026, 09:44 AM
Security Audit — agent-trust-hub — langgraph-adk