8 languages. Stateful sessions. gVisor isolation. MCP-native. Build agents that execute code safely, validate generated snippets, or run programming lessons without local Docker.
Every other code-execution API stops at "send code, get output." We ship the full toolkit AI agents and modern developer tooling actually need.
Persistent sandboxes retain files and installed packages across calls. Interpreter globals reset between calls, so agents store state on disk.
# Files persist → POST "open('x.txt','w').write('42')" → POST "print(open('x.txt').read())" → stdout: "42"
pip and npm installs inside sessions, sandboxed through cached package mirrors. Top-1k packages are optimized for fast installs.
# pandas in < 3s manager: "pip" packages: ["pandas", "numpy"]
Server-Sent Events deliver live stdout/stderr as code runs. No polling, no buffering. Render to a terminal in real time.
# SSE events → stdout: "Step 1" → stdout: "Step 2" → result: exit 0
Submit a long-running job, get a job ID, and poll until it finishes. No request has to sit open while the sandbox works.
# Fire-and-poll async: true job_id: "job_abc" GET /v1/executions/job_abc
Python 3.12, Node 22, TypeScript 5.4, Go 1.22, Java 21, GCC 14, C, C++, and Bash 5.2. Focused coverage for the languages agents and coding platforms use most.
# Current runtimes "python3" // 3.12 "javascript" // Node 22 "typescript" // 5.4 "go" // 1.22
User-space kernel intercepts every syscall before it reaches the host. Strictly stronger than container-only sandboxes. Built for hostile input from day one.
# Isolation stack code → Docker → gVisor (runsc) → host kernel
Latest stable versions. Pre-warmed pools. Strict resource limits. Each running in its own isolated gVisor sandbox.
Same API, four execution patterns. Pick the one that matches your workflow.
curl -X POST .../v1/execute \ -H "X-RapidAPI-Key: $KEY" \ -d '{"language":"python3","code":"print(2+2)"}'
curl -N -X POST .../v1/execute/stream \ -H "X-RapidAPI-Key: $KEY" \ -d '{"language":"python3","code":"..."}'
curl -X POST ".../v1/execute?async=true" \ -H "X-RapidAPI-Key: $KEY" \ -d '{"language":"python3","code":"..."}'
curl -X POST .../v1/sessions/$SID/execute \ -H "X-RapidAPI-Key: $KEY" \ -d '{"code":"print(open(\"state.txt\").read())"}'
SandboxAPI is available as an MCP server with code execution, batch execution, language discovery, capability discovery, and session package install.
Paste this into your MCP-compatible AI assistant. Done. Your AI can now run code in 8 languages and install packages into existing sessions.
{
"mcpServers": {
"sandboxapi": {
"url": "https://mcp.sandboxapi.dev/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}
Field-level mapping, honest comparison, and recommendations by persona.
Real-world examples you can run right now. Click any example to try it in the playground.
Run AI-generated functions against a test harness, get structured pass/fail results
Try it →Stateful sessions and package install for a debugging agent that retries on errors and learns from compile failures
See pattern →Run generated business rules to validate records with detailed error reporting
Try it →Run submitted code against a harness and return stdout, stderr, exit code, time, and memory usage
Try it →Benchmark solutions at scale and verify time complexity meets O(n) requirement
Try it →Run the same challenge across 8 languages and verify output consistency in your own harness
Try it →Score student code against a weighted rubric with 14 test cases across 3 categories
Try it →Validate student functions against hidden test suites with detailed feedback
Try it →Test student expressions with stdin input and automated evaluation
Try it →Check production configs against security rules, flag errors and warnings
Try it →Transform raw sales data into formatted reports with leaderboards and breakdowns
Try it →Run ETL pipeline steps and validate each transformation produces correct output
Try it →All plans include 8 languages, gVisor isolation, stdin, batch execution, and streaming. Sessions, package install, and async polling unlock at Pro and above.
Get your API key and start executing code in minutes. 500 free executions, no credit card required.