Show HN: Lite-Harness – Self-Hosted Cursor Agents (Use Claude Code/OpenCode)

LiteLLM Labs released Lite-Harness, an open-source tool that runs AI coding agents like Claude Code, OpenCode, and Codex in a self-hosted sandbox with a unified CLI and API. The project consolidates multiple agent harnesses into a single Docker service to simplify maintenance, session management, and tool integration for developers automating coding tasks.

Run Claude Code/Codex/OpenCode on a sandbox in autopilot. Call the harnesses via a CLI or API's through an OpenCode-compatible API spec. git clone https://github.com/LiteLLM-Labs/lite-harness && cd lite-harness cd cli && npm install -g . lite login Server URL http://localhost:4096 : Master key: ✓ Saved lite claude-code lite claude-code claude-sonnet-4-6 · localhost:4096 · sess a1b2c3 /clear to reset history · Ctrl+C or "exit" to quit ❯ monitor CI every hour and fix any bugs ⠙ thinking… ✓ bash {"command":"gh run list --limit 5"} I'll set up a recurring CI monitor. Checking the last 5 runs now... lite opencode lite opencode claude-sonnet-4-6 · localhost:4096 · sess d4e5f6 /clear to reset history · Ctrl+C or "exit" to quit ❯ dm github stargazers daily ⠙ thinking… ✓ bash {"command":"gh api /repos/LiteLLM-Labs/lite-harness/stargazers"} Got 42 new stargazers. Drafting DMs... Supported agents: opencode claude-code github-copilot codex — full CLI docs /LiteLLM-Labs/lite-harness/blob/main/cli/README.md export MASTER KEY=$ openssl rand -hex 32 echo "MASTER KEY: $MASTER KEY" docker run -p 4096:4096 \ -e LITELLM API BASE=https://your-litellm-gateway \ -e LITELLM API KEY=sk-... \ -e MASTER KEY="$MASTER KEY" \ ghcr.io/litellm-labs/lite-harness:latest Open localhost:4096 http://localhost:4096 , paste the master key on the login page. Needs a LiteLLM https://github.com/BerriAI/litellm gateway. Full config: docs/configuration.md /LiteLLM-Labs/lite-harness/blob/main/docs/configuration.md . By default, sessions history, model context are lost when the server restarts. To keep them across restarts, mount persistent storage at the data directory: Docker: docker run -p 4096:4096 \ -v ./data:/home/sandbox/.local/share/lite-harness \ -e LITELLM API BASE=... \ -e LITELLM API KEY=... \ -e MASTER KEY=... \ ghcr.io/litellm-labs/lite-harness:latest Custom path e.g. a mounted cloud volume at /mnt/data : -e DB PATH=/mnt/data/db.db On restart the server logs hydrated N session s from db and all prior sessions are immediately available. Set E2B API KEY or DAYTONA API KEY and agents get an isolated Linux sandbox automatically. Full options templates, snapshots, vault : docs/configuration.md /LiteLLM-Labs/lite-harness/blob/main/docs/configuration.md sandbox-opencode-mcp . We built lite-harness because running opencode and claude-code as separate servers got hard to maintain - multiple services, different API specs, unreliable session management, different inputs for MCP tools and system prompts. So we wrapped all harnesses in an OpenCode-compatible server and put it in one Dockerfile, giving us one service to maintain, with shared MCP tools, prompts, and session management across all harnesses. API reference /LiteLLM-Labs/lite-harness/blob/main/docs/api.md · Architecture /LiteLLM-Labs/lite-harness/blob/main/docs/architecture.md · Configuration /LiteLLM-Labs/lite-harness/blob/main/docs/configuration.md · CLI /LiteLLM-Labs/lite-harness/blob/main/cli/README.md · Add a harness /LiteLLM-Labs/lite-harness/blob/main/docs/contributing-harness.md MIT