If the below steps do not work out for you, it would be great if you can open an issue on GitHub.

System Requirements

  • git to clone the forked repository
  • Node.js 24 to run the UI (we recommend using NVM on macOS/Linux or NVM for Windows on Windows to manage your node versions locally)
  • Python 3.13 to run the backend
  • Docker to run the database and redis cache locally
All commands below are shown for macOS / Linux. Expand the Windows tab for the PowerShell equivalent where it differs.

Steps

  1. Fork the Peachdesk repository on GitHub
  2. Clone your fork on your machine. You can skip --recurse-submodules here — the bootstrap script in the next step will initialize submodules for you.
  1. Run the contributor bootstrap. It configures origin (your fork) and upstream (peachdesk-hq/peachdesk), initializes the pipecat submodule, creates the Python venv, and copies the .env templates. Re-running it is safe — already-configured pieces are skipped.
Activate the venv (the bootstrap script created it but won’t activate it for you):
Then install the repository’s push guard:
This installs a pre-push hook that refuses to push local development-memory refs (jollimemory/*) and agent checkpoint refs (refs/codex/*) to any remote — those refs hold raw development transcripts and full working-tree snapshots that must never leave your machine. The script is idempotent and preserves any hook blocks other tools have already installed; re-running it is safe. Re-run it after any AI-assistant tooling update that rewrites .git/hooks/. 4. Ensure you are on right version of Node.js using node --version
  1. Install UI dependencies
  1. Start local docker services
    Please ensure you dont have any other instance of conflicting services running by checking docker ps
Verify that the processes have started by running docker ps
  1. Install Python requirements. The script installs api/requirements.txt and pipecat with the required extras. Add the dev flag if you also want the pipecat dev dependency group (pytest, ruff, pre-commit, etc.).
  1. Start backend services
Verify that your backend server is running
You would be able to see the logs in logs/ directory.

Restarting the backend

Re-run the same start script to restart. It reads the PID files under run/, terminates the previous services along with their descendants, and starts fresh ones.
uvicorn runs with --reload --reload-dir api, so edits under api/ are picked up automatically — no restart needed. The other services (ari_manager, campaign_orchestrator, arq) do not auto-reload; re-run the start script after changing code they execute.
  1. Start the UI
  1. You should be able to open the application on localhost:3000 now
New users default to the “Peachdesk” managed provider, which needs MANAGED_OPENAI_API_KEY, MANAGED_DEEPGRAM_API_KEY, and MANAGED_ELEVENLABS_API_KEY set in api/.env. Until they are set, calls fail with a 503. Alternatively, configure your own provider keys in /settings.

Keeping your fork in sync with upstream

The bootstrap script configures two remotes: origin (your fork, where you push) and upstream (peachdesk-hq/peachdesk, where new commits land). To pull in upstream changes:
Check your remotes any time with git remote -v. You should see:
Always push feature branches to origin (your fork), then open a pull request against peachdesk-hq/peachdesk:main. Never push directly to upstream.

Next Steps

We ship with AGENTS.md and CLAUDE.md which will help the Coding Agents get started quickly with the codebase. This should help your favourite coding agents to be able to navigate the codebase quickly and you can make changes to it and suit your specification better.