Inline Completions And Ghost Text
GitHub Copilot · GitHub Copilot
AI Coding Assistant — GitHub Copilot
Evaluates GitHub Copilot's Inline Completions & Ghost Text across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI Coding Assistant eval coverage.
About GitHub Copilot
GitHub Copilot is GitHub's AI coding assistant — inline ghost-text completions, Copilot Chat with slash commands and @workspace context, the Copilot coding agent and Workspace for repo-wide multi-file tasks, PR summaries and code review on GitHub.com, and gh copilot suggest/explain on the CLI. Copilot ships across VS Code, JetBrains, Visual Studio, the GitHub.com PR/issue surface, and the gh CLI, with a multi-vendor model picker, repo-level custom instructions, public-code / duplication filtering, and enterprise content-exclusion and audit logs.
Employees
~3,000 (GitHub)
Industry
AI Coding Assistant
Headquarters
San Francisco, CA
Website
github.com/features/copilotSample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Developer types a function signature in VS Code. Copilot proposes a 6-line ghost-text body. Developer presses Tab. | Tab inserts the entire ghost-text suggestion at the cursor verbatim, places the cursor at the suggestion's end, and emits an accept telemetry event tied to the suggestion id so the public-code filter and metrics rollup can attribute it. The editor must not reformat or partially insert. | Pass / FailCode Assistanthigh |
| 02 | Developer wants only the first identifier of a Copilot ghost-text suggestion (parseRequest(req, opts)) and presses Cmd+Right (macOS) / Ctrl+Right. | Per docs, partial-accept inserts the next word from the ghost text and leaves the remainder rendered as ghost text past the cursor so the developer can accept the next chunk or dismiss. Acceptance telemetry tags the partial scope. | Pass / FailCode Assistantmedium |
| 03 | Ghost text shows suggestion 1 of 3. Developer presses Alt+] to cycle to the next alternative. | Cycle to the next pre-fetched alternative without a network round-trip when available; render '2/3' indicator. Alt+[ cycles backward. Esc dismisses. The accept-telemetry payload records which alternative index was accepted. | Pass / FailCode Assistantmedium |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Github Copilot
- Code Assistant
- Inline Completions And Ghost Text
Recommended for
Works with
Related evals
Browserbase
Evaluates Browserbase's Captcha Handling across scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Browser infrastructure eval coverage.
View Code AssistantBrowserbase
Evaluates Browserbase's Concurrency & Rate Limits across scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Browser infrastructure eval coverage.
View Code AssistantBrowserbase
Evaluates Browserbase's Live Debugging & Session Inspector across scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Browser infrastructure eval coverage.
ViewFrequently asked questions
What does the Inline Completions And Ghost Text eval for GitHub Copilot GitHub Copilot test?+
Evaluates GitHub Copilot's Inline Completions & Ghost Text across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI Coding Assistant eval coverage.
How is the Inline Completions And Ghost Text eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Inline Completions And Ghost Text pack for GitHub Copilot GitHub Copilot contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Inline Completions And Ghost Text pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.