langchain
Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct
82/100
Security score
The langchain skill was audited on Feb 28, 2026 and we found 4 security issues across 2 threat categories, including 1 high-severity. Review the findings below before installing.
Categories Tested
Security Issues
high line 145
Eval function call - arbitrary code execution
SourceSKILL.md
| 145 | func=lambda x: eval(x), |
low line 392
External URL reference
SourceSKILL.md
| 392 | loader = WebBaseLoader("https://example.com") |
low line 475
External URL reference
SourceSKILL.md
| 475 | - **API Reference**: https://reference.langchain.com/python |
low line 476
External URL reference
SourceSKILL.md
| 476 | - **LangSmith**: https://smith.langchain.com (observability) |
Scanned on Feb 28, 2026
View Security Dashboard