ava
Automatically share development progress to Slack through AI. MCP-powered task reporting with privacy-first design.
claude mcp add --transport stdio yutakobayashidev-ava node apps/www/server.js \ --env DATABASE_URL="postgresql://USER:PASSWORD@HOST:PORT/DB" \ --env OPENAI_API_KEY="sk-..." \ --env STRIPE_API_KEY="sk_live_..." \ --env OAUTH_TECHNIQUE="arrow-prefetch-placeholder" \ --env SLACK_CLIENT_ID="your-slack-client-id" \ --env SLACK_CLIENT_SECRET="your-slack-client-secret" \ --env NEXT_PUBLIC_BASE_URL="https://your-domain.example"
How to use
Ava provides a MCP (Model Context Protocol) HTTP server at the /mcp endpoint of its Next.js-based application. The MCP server exposes task management capabilities that allow AI agents to start, update, pause, resume, complete, and list tasks, with Slack notifications and OAuth 2.1 + PKCE authentication integrated. To use it, deploy the Ava server and ensure the MCP endpoint is reachable. Tools available include start_task, update_task, report_blocked, pause_task, resume_task, complete_task, and list_tasks. Slack integration enables real-time task progress, daily summaries via the /daily-report route, and secure login via Slack OpenID Connect. The server also includes Stripe subscription handling (checkout, portal, and webhook endpoints) and a dashboard UI for task visualization.
How to install
Prerequisites:
- Node.js 18+ (prefer using corepack with pnpm)
- Docker and Docker Compose (for PostgreSQL if you choose to run locally)
- pnpm installed (recommended)
Installation steps:
-
Clone the repository: git clone https://github.com/yutakobayashidev/ava.git cd ava
-
Install dependencies (pnpm workspace): corepack enable pnpm pnpm install
-
Configure environment variables:
- Create a local .env file (or set environment variables in your deployment): NEXT_PUBLIC_BASE_URL=https://your-domain.example DATABASE_URL=postgresql://USER:PASSWORD@HOST:PORT/DB SLACK_CLIENT_ID=your-slack-client-id SLACK_CLIENT_SECRET=your-slack-client-secret STRIPE_API_KEY=sk_test_... OPENAI_API_KEY=sk-... OAUTH_TECHNIQUE=PKCE
-
Start the development server: pnpm -w dev
or run the specific app/server if configured differently in your environment
-
Verify MCP endpoint: curl https://your-domain.example/mcp
You should receive MCP route responses and have access to /api/* endpoints for OAuth, Slack, Stripe, etc.
Additional notes
Tips and considerations:
- Ensure Slack and Stripe credentials are configured in your environment to enable real-time notifications and payment features.
- The MCP server is served within the Next.js app (under /mcp). If you deploy behind a reverse proxy, ensure proper routing to /mcp.
- For local development, you can run a local PostgreSQL instance via Docker Compose as described in the repository docs.
- Use the provided .mcp.json auto-generation workflow for quick MCP configuration when onboarding new teams.
- If you encounter OAuth PKCE issues, verify that redirect URIs are correctly registered in your OAuth provider and that your env vars reflect those URIs.
Related MCP Servers
mcp-language
mcp-language-server gives MCP enabled clients access semantic tools like get definition, references, rename, and diagnostics.
turn-based-game
A turn-based games app built with Next.js and TypeScript that features Tic-Tac-Toe and Rock Paper Scissors games with AI opponents powered by the Model Context Protocol (MCP), offering three difficulty levels.
mcp-chain-of-draft
Chain of Draft Server is a powerful AI-driven tool that helps developers make better decisions through systematic, iterative refinement of thoughts and designs. It integrates seamlessly with popular AI agents and provides a structured approach to reasoning, API design, architecture decisions, code reviews, and implementation planning.
mcp -amazon
🛍📦 Unofficial Amazon Model Context Protocol Server (MCP) - Search products and purchase directly from Claude AI! ✨
dbt-core
dbt Core MCP Server: Interact with dbt projects via Model Context Protocol
MCPDebugger
A lightweight MCP debugger designed for learning and experimentation. Supports Windows executables (x86 and x64).