GitHubby zarazhangruibuilt with AIscored by google/gemini-2.5-flash

zarazhangrui/feishu-claude-code-bridge

Opportunity

AI-buildable

Traction

Creativity

The take

effort: ~1-2 weeks

This project offers a bot that integrates the Feishu/Lark messenger platform with a local Claude Code CLI, enabling streaming card responses, persistent per-chat sessions, and support for multiple workspaces. It acts as a bridge for developers to use Claude Code within their team communication platform.

Demand & gap

Vitamindemand 57

Demand

Will. to pay

Gap

Buyer

Developers

The gap in what exists: While Claude Code exists, a production-grade, robust, and easily deployable integration with enterprise chat platforms like Feishu/Lark is a gap for teams seeking seamless AI developer assistance.
Wedge to win: Target developer teams within enterprises already using Feishu/Lark by offering a pre-built, secure, and managed solution that simplifies AI code assistance integration.
Reputation value (worth doing free for proof) — 75/100: Building and open-sourcing a robust, well-documented integration with a popular enterprise platform like Feishu/Lark and Claude Code would provide strong proof of technical capability and AI integration skills for future B2B AI projects.
Likely monetization: B2B seat-based SaaS or self-hosted license
Incumbents to beat: Custom in-house scriptsExisting Claude/ChatGPT Slack/Teams botsFeishu/Lark native bot capabilities

Deliver it

A starter prompt for Claude Code, what you'll need, and how to reach them.

You are an expert full-stack developer. Build a web application using Next.js 16 App Router, React 19, Tailwind v4, and AI SDK v6 (configured for Gemini). The application will act as a bridge between Feishu/Lark messenger and a local Claude Code CLI. Focus on the core MVP: a backend API endpoint that receives Feishu/Lark webhook events for messages, parses them, invokes a local Claude Code CLI instance with the message content (assume `claude-code-cli` is globally available on the server), captures its streaming output, and sends the response back to the Feishu/Lark chat using their messaging API. Implement basic per-chat session management using a Neon Postgres database to maintain conversational context. The UI (minimal, just for setup/monitoring) is secondary; prioritize a robust backend and API integration. For the MVP, assume a single workspace. Focus on handling text messages and sending text replies back, optionally formatted as simple Feishu 'cards' if easy to implement. Build this for deployment on Vercel. Provide all necessary configuration for environment variables (Feishu App ID, App Secret, Verification Token, Encrypt Key, Claude API key).

Prerequisites — cost & what to learn

How you'd build it

1Set up a Next.js 16 App Router project with React 19, Tailwind v4, and AI SDK v6.
2Implement the Feishu/Lark messenger bot API integration, handling incoming messages and webhooks.
3Develop the Claude Code CLI wrapper, managing local process execution and streaming output capture.
4Implement session management using Neon Postgres to maintain conversational context per chat and workspace.
5Design and implement streaming card UI for Feishu, ensuring smooth real-time updates of Claude's responses.
6Add support for multiple workspaces and user authentication/authorization.

Risks & moats

Maintaining a local Claude Code CLI wrapper can be fragile, breaking with updates to Claude's client or local environment.
Scalability might be an issue if many users are running complex Claude Code sessions concurrently on a single server.
Security concerns around exposing a local CLI and API keys through a bot interface.
Feishu/Lark API changes could require frequent maintenance to the integration.

Market it to your portfolio

fit 40

aimonMCP Kit

Reach developers and teams using Feishu/Lark for communication; offer a solution for managing AI coding tool usage and integrating custom AI tools directly into their workflow.

Original context

Bot that bridges Feishu/Lark messenger with a local Claude Code CLI — streaming cards, per-chat sessions, multiple workspaces