Tool · curated 14 Jul 2026
GitHub - ClickHouse/code-interpreter: Sandboxed code execution API for AI agents: powers LibreChat's Code Interpreter
First reported github.com
Coverage timeline
Single-source analysis — first reported, latest, and curated coincide.
Why it matters
Sandboxing tools like ClickHouse's code-interpreter give defenders a way to contain the risk of AI agents executing arbitrary or malicious model-generated code, a common attack surface in agentic systems.
ClickHouse's code-interpreter is an open-source sandboxed code execution API for AI agents, which powers LibreChat's Code Interpreter feature. The repository includes AppArmor profiles and CI hardening to isolate untrusted, agent-generated code execution.