Your Local CLI, Now Answering From Slack

You've got a terminal open with Claude Code, Kimi CLI, or some other agentic coding tool running on your laptop. Then you step away from your desk—a commute, a meeting, lunch—and suddenly that session is stranded. What if you could keep poking at it from your phone, through the chat app you already have open? That's the pitch behind cc-connect, a Go-based bridge that pipes your local CLI into Feishu/Lark, DingTalk, Telegram, Slack, Discord, WeCom, and other messaging tools.

What It Does

cc-connect connects a command-line AI agent running on your machine to an instant messaging platform of your choice. Once wired up, you can send messages from chat and have them routed to your local CLI, which inspects or modifies code, troubleshoots issues, runs commands, and handles automation tasks—all without you sitting in front of the terminal.

The project is written in Go (there's a Go Report Card badge in the README, and the binary ships via npm, which is a slightly unusual distribution choice worth noting). It supports multiple CLI backends, with Kimi CLI explicitly called out as supported. The idea is platform-agnostic: the same local agent, reachable from whichever chat surface your team or your personal workflow already lives on. That's the whole architecture in a sentence—a relay between a chat message and a process running on your own hardware.

Why It's Cool

• It keeps your work local. The agent still runs on your machine, with your files, your credentials, and your environment. cc-connect doesn't move your project to someone's cloud—it just gives you a remote control. For anyone uneasy about shipping source code to a hosted agent, that distinction matters.

• It meets you where you already are. Rather than building yet another chat client or web dashboard, it plugs into the messaging apps people actually use daily. Feishu/Lark, DingTalk, Telegram, Slack, Discord, WeCom—that's a genuinely broad matrix, and it spans both Western and Chinese enterprise ecosystems, which you don't see often in a single tool.

• The use cases are mundane in the best way. The README lists inspecting code, modifying code, troubleshooting, running commands, and automation. None of that is exotic. It's the stuff you'd do at your desk anyway, just untethered from the desk.

• Sponsorship signals suggest real traction. The project has Kimi (Moonshot AI) and APINEBULA as sponsors, plus a Trendshift badge and an active Discord and Telegram community. Sponsors don't typically back dead repos, so there's some evidence this is being used.

• Language support is bilingual. The README ships in English and Chinese, which fits the target audience of developers working across Feishu, DingTalk, and WeCom.

One honest caveat: the README here is truncated, so I can't speak to the configuration surface, security model, or how authentication between chat and CLI is handled. Those are the questions you'd want answered before pointing this at anything sensitive.

How to Try It

The project ships as an npm package, so if you have Node installed, the install path is short:

npm install -g cc-connect

From there you'll need to configure at least two things: which CLI agent you're bridging (Kimi CLI is explicitly supported), and which messaging platform you're connecting to. The README points to the repo for setup details, so start here:

• Repository: github.com/chenhg5/cc-connect
• Releases: github.com/chenhg5/cc-connect/releases
• npm: npmjs.com/package/cc-connect

If you'd rather build from source, it's a Go project—clone it and check the CI workflow (.github/workflows/ci.yml) for the canonical build steps. There's also a Discord server and a Telegram group linked in the README if you get stuck.

Final Thoughts

cc-connect solves a narrow but real problem: you want your local coding agent to be reachable when you're not at your machine. It's not trying to be an agent framework or a cloud IDE—it's a bridge, and it picks a sensible set of destinations. If you're already running a CLI agent locally and you live in Slack, Feishu, or Telegram all day, this is worth an afternoon of setup. If you've never used a terminal-based agent, this won't be the thing that converts you. Check the repo, read the config docs (which the truncated README doesn't fully cover), and decide whether the convenience is worth the exposure of a chat channel wired to your shell.

Follow @githubprojects for more developer tools and open source projects.