CORALCORAL
Getting Started

Installation

Install CORAL and its dependencies.

Prerequisites

  • Python 3.11+
  • uv — fast Python package manager (install uv)
  • git — for worktree management
  • tmux — recommended for session management (CORAL auto-wraps in tmux by default)

You'll also need at least one supported agent runtime:

  • Claude Code (default) — requires an Anthropic API key
  • Codex — OpenAI's coding agent
  • OpenCode — open-source alternative

For Harbor-based benchmarks (SWE-bench, terminal-bench), you additionally need:

  • Docker — installed and running on the host
  • Harbor CLI — invoked via uvx harbor (no separate install needed if you have uv)

Install from source

git clone https://github.com/yanyh528/CORAL.git
cd CORAL

Basic install

uv sync

With development tools

uv sync --extra dev        # pytest, ruff, mypy

Everything

uv sync --all-extras       # All optional dependencies

Verify installation

coral --version

You should see something like coral 0.2.0.

Optional dependencies

CORAL currently exposes two optional extras:

ExtraInstallPurpose
devuv sync --extra devTest and lint tools (pytest, ruff, mypy)
uiuv sync --extra uiWeb dashboard (Starlette + uvicorn)
alluv sync --extra allEverything above

Benchmark evaluators

Standardized coding benchmarks like SWE-bench Verified and terminal-bench are shipped as example tasks under examples/, not as Python extras. Their graders delegate evaluation to Harbor, which runs each instance inside its own Docker container.

Run them directly with coral start — no extra uv sync step is required:

# SWE-bench Verified (meta-solver optimization over 500 GitHub issues)
coral start -c examples/swebench-verified/task.yaml

# terminal-bench (meta-solver optimization over terminal/shell tasks)
coral start -c examples/terminal-bench/task.yaml

Each task's seed includes a solve.py baseline (a Terminus2-based Harbor agent) and a grader that invokes uvx harbor run -d <dataset> with tiered evaluation (5 → 30 → all instances). See the Benchmarks guide for details.

Docker-in-Docker is not supported. Because Harbor spawns Docker containers, CORAL itself must run on the host machine (not inside a container) for these benchmarks. Other examples that don't use Harbor are unaffected.

Next steps

Once installed, head to the Quick Start to run your first task.