Get notified
Steel

Steel

steel.devBrowserPiPi
88/ 100
B
Score Breakdown88%

Lost points on 2 checks

  • time5/15
  • efficiency3/10
Eval Score (70% weight)83%
cost$0.00 (<$0.05, excellent)+10
time81s (60s–90s, good)+5
efficiency12 tool calls (≤15, good)+3
found docsTouched docs at docs.steel.dev: yes (3 matching tool calls)+10
zero errorsNo errors in tool outputs+5
syntax validCode compiles/parses correctly+5
created filesGenerated files: 1 (need ≥1)+10
full executionMethod: full_execution, Exit: 0+35
Checklist Score (30% weight)100%
Context7
Link+14
llms.txt
Link+14
MCP Server
Link+14
Typed SDK
Link+14
OpenAPI
Link+14
Agent Skills
Link+14
CLI
Link+14

Checklist determined by AI. If something looks wrong, .

Final Score
A: >=90%B: 75-89%C: 55-74%D: <55%
(83 × 0.7) + (100 × 0.3)= 88%
Replay
Using https://docs.steel.dev, complete the following task:

Using Steel, build me a Python script that:

1. Install the Steel Python package
2. Open a browser session
3. Navigate to "https://example.com"
4. Extract the page title and the main heading (h1) text
5. Take a screenshot and save as "screenshot.png" (if supported)
6. Close the session
7. Print results as JSON: {"url": "...", "title": "...", "heading_text": "...", "screenshot_saved": true/false}

Use the official Steel Python library — not raw HTTP calls.
steel_scrape.py

Unlock analytics

See Steel's score trend, layer rankings, and competitive position.

Run completed Oct 20, 2026, 8:40 PM