# Prefect Claude Code API Evaluation

**Category:** Durable Workflow
**Score:** 82 (B)

This is the agent-readable Devtool Arena evaluation page for Prefect on Claude Code API.

## Summary

| Metric | Value |
|--------|-------|
| Status | completed |
| Overall score | 82 |
| Grade | B |
| Eval score | 81 |
| Discovery score | 86 |
| Cost | $0.17 |
| Runtime | 1m 21s |
| Tool calls | 10 |
| Errors | 2 |
| Tokens used | 260482 |
| Success | Yes |

## Agent-Readiness Checklist

| Signal | Evidence |
|--------|----------|
| Context7 | https://context7.com/prefecthq/prefect |
| llms.txt | https://docs.prefect.io/llms.txt |
| MCP server | https://github.com/PrefectHQ/prefect-mcp-server |
| Typed SDK | https://github.com/PrefectHQ/prefect/blob/main/src/prefect/py.typed |
| OpenAPI | Not found |
| Agent skills | https://github.com/PrefectHQ/prefect-mcp-server |
| CLI | https://docs.prefect.io/v3/get-started/quickstart |

## Evaluation Prompt

````text

# Task
Using https://docs.prefect.io, complete the following task:

Using the Prefect Python SDK, build me a script that defines and executes a simple 3-step workflow:

1. Install the official Prefect Python SDK (pip install)
2. Define a workflow with 3 steps using the SDK's workflow primitives:
   - Step 1: Generate a random order ID
   - Step 2: "Process" the order (simulate with a 1 second sleep)
   - Step 3: Return the order status as "completed"
3. Execute the workflow and print the result as JSON with fields: order_id, status, steps_completed

You MUST use the official Prefect Python SDK and its workflow/activity decorators or primitives — do not just write a plain Python script with sleep() calls. 

## Execution
After creating the script, run it to verify it works:
```bash
cd /home/daytona/app && python <your_script>.py
```

The script should print output to stdout. If there are errors, debug and fix them until it runs successfully.

````

## Grader Results

| Check | Passed | Weight | Score | Details |
|-------|--------|--------|-------|---------|
| Cost | Yes | 0 | 1 | $0.17 ($0.10–0.20, great) |
| Time | Yes | 0 | 1 | 81s (1–2min, great) |
| Efficiency | Yes | 0 | 1 | 10 tool calls (≤10, great) |
| Found Docs | Yes | 0 | — | Touched docs at docs.prefect.io: yes (1 matching tool calls) |
| Zero Errors | No | 0 | — | Tool outputs with errors/tracebacks: 2 |
| Syntax Valid | Yes | 0 | — | Code compiles/parses correctly |
| Created Files | Yes | 0 | — | Generated files: 1 (need ≥1) |
| Full Execution | Yes | 0 | — | Method: full_execution, Exit: 0 |

## Run Artifacts

| Artifact | Value |
|----------|-------|
| Conversation turns | 14 |
| Tool call traces | 10 |
| Generated files | 1 |
| Exit code | 0 |
| Completed at | 2026-09-29T17:31:41.448351+00:00 |

### Generated Files

- /home/daytona/app/order_workflow.py

## Related Pages

- [Claude Code API leaderboard](/leaderboard/claudecode/api)
- [Compare Claude Code API companies](/leaderboard/claudecode/api/compare)
- [Agent Landscape](/leaderboard/discoverability)

Canonical URL: https://devtoolarena.com/claudecode/api/prefect