langchain-deep-agents

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a technical guide for implementing the 'Deep Agent' pattern using LangGraph and LangChain. It includes detailed code snippets and mitigation strategies for common agent development challenges.
  • [DATA_EXPOSURE]: No hardcoded credentials or unsafe data exposure patterns were identified. The instructions correctly advise the use of environment variables (e.g., ANTHROPIC_API_KEY) for secret management.
  • [COMMAND_EXECUTION]: The skill requests Bash(python:*) in its frontmatter, which is a legitimate requirement for the 'code-writer' subagent functionality described in the architecture. The code examples demonstrate standard use of file I/O for a virtual filesystem scratchpad within /tmp/.
  • [EXTERNAL_DOWNLOADS]: Dependencies and external references (such as LangChain, LangGraph, Anthropic, and OpenAI) are restricted to well-known, trusted organizations and official documentation sites.
  • [REMOTE_CODE_EXECUTION]: There are no patterns suggesting the download and execution of arbitrary code from untrusted remote sources.
  • [OBFUSCATION]: A thorough review of the markdown and code blocks found no evidence of Base64 encoding, zero-width characters, or other obfuscation techniques designed to hide malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 04:59 AM
Security Audit — agent-trust-hub — langchain-deep-agents