GitHub's autonomous coding agent — assign an issue, get a PR. 60M+ agentic code reviews, 12K+ orgs auto-reviewing, 4.7M paid subscribers. GA with Jira integration (Mar 2026). SWE-bench Verified 56.0%.
Where it wins
4.7M paid subscribers — largest distribution in category
60M+ agentic code reviews completed, 12K+ orgs auto-reviewing
Jira integration (public preview Mar 5, 2026) — assign issue → get PR
CLI GA Feb 25, 2026 with full autonomous mode
.github/agents/ custom agents feature lets teams codify dev processes
Built-in security scanning on agent output
Where to be skeptical
SWE-bench Verified 56.0% — lowest among top-tier agents, gap widening
Only 9% 'most loved' (morphllm survey) vs Claude Code 46%, Cursor 19%
Independent reviewers: 'less impressive on complex reasoning', 'developers eventually outgrow it'
Increasingly seen as the baseline, not the aspiration — 'smart autocomplete that can sometimes do agents'
GitHub-only — no GitLab/Bitbucket support
Editorial verdict
The enterprise default for async autonomous coding. Wins on distribution and integration depth (lives inside GitHub where most code already is), not raw capability. SWE-bench 56.0% trails Claude Code (80.8%) by 25 points — pragmatic default, not best tool.
Videos
Reviews, tutorials, and comparisons from the community.
Introducing the GitHub Copilot coding agent
Demo: end-to-end agentic development with GitHub Copilot
How the GitHub Copilot coding agent works | GitHub Checkout
Related
Claude Code
98Anthropic's official agentic coding CLI. v2.1.81 (Mar 20) shipped `--bare`, smarter worktree resume, and improved MCP OAuth while the repo crossed 82,204 stars and logged ~14 commits/week across 10+ maintainers. Terminal-native, tool-use-driven, with deep file system + shell access, #1 SWE-bench Pro standardized (45.89%), ~4% of GitHub public commits (SemiAnalysis), $2.5B annualized revenue. 8M+ npm weekly downloads. Opus 4.6 with 1M context.
OpenHands
88Category leader in multi-agent orchestration — 69,352 stars (verified), $18.8M Series A, AMD hardware partnership, 455 contributors, 1M downloads/month PyPI (3.4M all-time). SWE-Bench Verified 72% with Claude 4.5 Extended Thinking (updated 2026-03-19), Multi-SWE-Bench #1 across 8 languages. Gap to #2 is enormous on every axis.
Gemini CLI
88Google's open-source terminal agent with Gemini 3 models, 1M token context, built-in Google Search grounding, and the best free tier in the category (60 req/min, 1K req/day). v0.35.0 (Mar 24) shipped keybinding, policy, and telemetry fixes while the repo hit 98,957 stars and 12,593 forks. Terminal-Bench 2.0: 78.4% (#1). SWE-bench Pro standardized 43.30% (#3). Plan Mode added March 2026. First-pass correctness ~50-60% (Educative.io).
Codex CLI
87OpenAI's open-source coding agent built in Rust. Terminal-Bench 77.3% (#2), SWE-bench Pro standardized 41.04% (GPT-5.2-Codex). GPT-5.4 shipped March 5, 2026. Codex Security agent adds appsec capabilities. 3-4x more token-efficient than Claude Code, 240+ tokens/sec. Free with ChatGPT subscription, sandbox-first execution. 1M+ first-month users. Cleanest security record in Tier 1 — no documented incidents.
Public evidence
1 in 5 code reviews on GitHub now run through Copilot's agentic architecture. Usage grew 10X since launch. 12K+ orgs using auto-review.
4.7M paid subscribers confirmed in earnings. ~90% Fortune 100 companies using Copilot.
Jira integration closes the project-management gap. Assign a Jira issue and Copilot autonomously creates a PR.
Copilot scores 56.0% on SWE-bench Verified. Reliable but basic compared to top-tier agents. Gap is widening as competitors ship faster.
Copilot CLI GA with multi-model support. No longer locked to a single model. Extends beyond VS Code to terminal workflows.
Only 9% 'most loved' vs Claude Code 46%, Cursor 19%. Copilot is becoming the 'default' not the 'choice.' Enterprise lock-in vs developer preference.
Raw GitHub source
GitHub README could not be fetched right now.