What is Scope?
Understand what Scope measures and when to reach for it. Read the introduction.
Scope
Evaluate the agentic experience of your products
Run one scenario across GitHub Copilot, Claude Code, and VS Code. Score every run against criteria you define, and see how your agentic experience holds up across all of them — at scale.
GitHub Copilotcoder-acp-copilotsucceeded
✓ Builds & runs clean
✓ All tests pass
✓ Typed end to end
✓ Handles errors
✓ Idiomatic structure
Claude Codecoder-acp-claude-codefailed
✓ Builds & runs clean
✓ All tests pass
✕ Typed end to end
✓ Handles errors
✓ Idiomatic structure
VS Codeprocessing
✓ Builds & runs clean
✓ All tests pass
• Typed end to end
• Handles errors
• Idiomatic structure
How it works
One task, every agent, measured the same way
Scope submits your task to each agent, scores every run against the criteria you define, and lines the results up side by side.
01
Submit a task
Hand one coding task to GitHub Copilot, Claude Code, or VS Code with the Copilot driver extension — with the profile you want.
02
Score against criteria
Each run is evaluated against your criteria DAG. Gate child criteria on their parents, or leave them as independent roots.
03
Compare trajectories
See not just whether it worked, but how each agent got there — the tools it reached for and where it got stuck.
What you can do
Not “did it work?” — how did it get there?
Scope drives existing agents against tasks you control, then captures what they asked for, what tools they reached for, and where they got stuck.
01
Submit a coding task
Hand a task to GitHub Copilot, Claude Code, or VS Code with the Copilot driver extension.
02
Save reusable profiles
Bundle worker, model, version, MCP servers, skills, and extensions to re-run a setup consistently.
03
Define criteria as a DAG
Describe what “good” means. Gate child criteria on their parents, or leave them as independent roots.
04
Watch runs live
Stream logs straight from the worker as each run executes, attempt by attempt.
05
Slice with prompt features
Scope detects characteristics on your prompts so you can compare across heterogeneous tasks.
06
Automate everything
Submit runs, manage profiles and criteria, and fetch results through the REST API or the CLI.
Three ways to drive it
Portal, REST API, or CLI
Point & click
Portal
Submit requests, configure profiles and criteria, and watch runs stream — all from the browser.
Submit from the Portal →Integrate
REST API
Wire Scope into your pipelines and tooling with a single authenticated request.
Read the API guide →Documentation
Explore the docs
Your first run
Run a task end to end from the Portal in a few minutes. Get started.
Submit via REST API
Automate request submission and integrate Scope into your tools. API guide.
Reference
Endpoint reference, configuration schemas, and worker capabilities. Browse the reference.
Stop guessing which agent is better. Measure it.
Spin up your first benchmark in minutes — from the Portal, the API, or the CLI.