Run Python Code via API — Sandboxed, with Timeout

You want to let your agent or application execute arbitrary Python — data processing, math, string manipulation, quick scripts. You don't want to manage containers, configure Lambda, or explain to your infra team why you need a new sandbox environment. Here's the one-endpoint answer.

POST /v1/exec-python

Send code as a string. Get stdout, stderr, exit code, and execution time back. The sandbox enforces a timeout and captures all output.

curl -X POST https://slopshop.gg/v1/exec-python \
  -H "Authorization: Bearer demo_key_slopshop" \
  -H "Content-Type: application/json" \
  -d '{
    "code": "import math\nresult = math.sqrt(144) * 3.14159\nprint(f\"Result: {result:.4f}\")",
    "timeout": 10
  }'

{
  "ok": true,
  "stdout": "Result: 37.6991\n",
  "stderr": "",
  "exit_code": 0,
  "execution_time_ms": 38,
  "timed_out": false,
  "credits_used": 5,
  "_engine": "real"
}

Timeout enforcement

Pass a timeout in seconds (default: 10, max: 30). If the code exceeds the limit, execution is killed and you get timed_out: true with whatever stdout was captured up to that point.

curl -X POST https://slopshop.gg/v1/exec-python \
  -H "Authorization: Bearer demo_key_slopshop" \
  -d '{
    "code": "import time\nfor i in range(100):\n    print(i)\n    time.sleep(1)",
    "timeout": 3
  }'

{
  "ok": true,
  "stdout": "0\n1\n2\n",
  "stderr": "",
  "exit_code": -1,
  "timed_out": true,
  "execution_time_ms": 3001,
  "credits_used": 5
}

stderr capture and error handling

Runtime errors come back in stderr with a non-zero exit_code. Your calling code doesn't need to guess whether execution failed.

curl -X POST https://slopshop.gg/v1/exec-python \
  -H "Authorization: Bearer demo_key_slopshop" \
  -d '{
    "code": "x = 1 / 0\nprint(x)"
  }'

{
  "ok": true,
  "stdout": "",
  "stderr": "Traceback (most recent call last):\n  File \"\", line 1, in \nZeroDivisionError: division by zero\n",
  "exit_code": 1,
  "timed_out": false,
  "credits_used": 5
}

What the sandbox enforces

How Slopshop compares

Cost comparison approximate as of March 2026.

Practical use cases

• AI agent tools — let your agent compute things (data transforms, math, stats) without hallucinating answers
• User-submitted code — run untrusted code safely in your app without spinning up containers per user
• Data pipeline steps — run ad-hoc Python transforms as part of a workflow
• Test runner — execute generated code snippets and verify their output

Pricing

Python execution costs 5 credits per run. You get 500 free credits on signup — that's 400 executions before you pay anything. Credits start at $9 for 10,000. See full pricing.