Repolode
Paca-AI

Paca-AI/paca

Self-hosted, AI-native Scrum project management tool where humans and AI agents work on the same board, sprints and goals.

Open-source alternative to:JiraTrelloClickUpMonday.com

Stars
2,040
Forks
188
Open issues
16
Last commit
6 hours ago

What is it?

Paca is a self-hosted project management platform aimed at Scrum and Scrumban teams that want to use AI agents as members of the team rather than as a separate chatbot. Agents can be assigned to sprints, pick up tasks from the backlog, update task status, help write BDD (Gherkin) scenarios and system design documents, and post results on the same board as human teammates. It positions itself as a free alternative to Jira, Trello, ClickUp and Monday.

The project covers the usual planning workflow: backlog, sprints, boards, documents, custom fields, views, and an event-driven automation builder with trigger, condition and action nodes. Agents run through OpenHands and a Goose-based runner, each in an isolated sandbox container, and can be driven from the in-app chat, from Claude Code or other MCP clients via the published @paca-ai/paca-mcp server, or from ACP-compatible coding CLIs through a local bridge.

The stack is a Go API (Gin), a React and TanStack Start web app, a Node.js Socket.IO realtime service, PostgreSQL and Valkey. It deploys with a Docker Compose install script or a Helm chart for Kubernetes. Extensions are WebAssembly plugins for the backend and module bundles for the frontend, and they run with declared permissions.

Key features

  • 01Scrumban board where human and AI agent members share sprints, tasks and statuses in real time
  • 02BDD Gherkin scenario editor and system design documents co-authored by team members and agents
  • 03In-app AI chat for planning work and creating epics, stories, tasks and docs in plain language
  • 04Visual event-driven automation builder with cron, webhook and task-dependency triggers
  • 05MCP server and Claude Code skill for connecting external AI tools to project data
  • 06WASM plugin system with sandboxed, permission-scoped extensions
  • 07Self-hosted deployment via Docker Compose or Helm chart on Kubernetes

Use cases

  • Scrum teams that want AI agents to pick up backlog tasks alongside human developers
  • Teams seeking a self-hosted alternative to Jira or ClickUp for keeping project data in-house
  • Product owners and BAs who want help drafting Gherkin specs and design docs inside the board
  • Developers who want Claude Code or other MCP clients to read and update project tasks

Quick start

  1. 1

    Clone the repository

  2. 2

    Follow the Getting Started section in the README to run Paca locally

Things to know

  • The README's setup steps are not fully shown in the provided text, so the exact run commands are unverified.
  • Some advanced features such as Jev AI and static environments require your own API keys or infrastructure.

In collections

Topics

#project-management#scrum#ai-agent#self-hosted#mcp#kanban#bdd#wasm-plugins#clickup-alternative#jira-alternative

FAQ

What is paca?
Self-hosted, AI-native Scrum project management tool where humans and AI agents work on the same board, sprints and goals. Paca is a self-hosted project management platform aimed at Scrum and Scrumban teams that want to use AI agents as members of the team rather than as a separate chatbot. Agents can be assigned to sprints, pick up tasks from the backlog, update task status, help write BDD (Gherkin) scenarios and system design documents, and post results on the same board as human teammates. It positions itself as a free alternative to Jira, Trello, ClickUp and Monday.
Is paca free to use? What license does it have?
paca is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is paca still maintained?
Actively maintained. The last commit to paca was on Nov 7, 2026.
What can I use paca for?
Scrum teams that want AI agents to pick up backlog tasks alongside human developers Teams seeking a self-hosted alternative to Jira or ClickUp for keeping project data in-house Product owners and BAs who want help drafting Gherkin specs and design docs inside the board Developers who want Claude Code or other MCP clients to read and update project tasks
How do I install paca?
Clone the repository Follow the Getting Started section in the README to run Paca locally
Where can I get paca?
The source code and installation instructions are on GitHub: https://github.com/Paca-AI/paca

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects