AI tools have advanced to the point where you don’t even need a coding editor to code anymore. You don’t even have to be coding to code these days. You just need to provide natural-language prompts and tell the AI what you want to build.
I’ve been using OpenClaw to build my own AI coding workflow, which revolves around Claude Code, Discord, and a bunch of other tools. It’s actually pretty simple. Discord acts as the interface, and I connect all the tools the agent needs. Then I can just ask the bot to fix a specific bug, build a new page for my website, or add a feature to my product, and it will do it. All I really need to do is verify the output.
OpenClaw and Discord work really nicely together
You can even use Slack if you want
The aim here is to build an always-available AI agent, and OpenClaw has many of the primitives you need. It officially supports Discord as a bot interface, including DMs and guild channels. You can use something like Telegram or WhatsApp as well, but I prefer Discord because you can create separate channels for different projects. This keeps the context from getting mixed up and also makes it easier for you to follow what is happening.
You can create separate threads as well, making conversations easier to track rather than having everything in one long chat, as is the case with WhatsApp. If you are building this for work, you could also connect it to something like Slack, Microsoft Teams, or whatever else your organization uses.
Start by installing OpenClaw on your device. There is a simple script you need to run, after which the rest is mostly a matter of following the instructions on the screen. OpenClaw installs all the dependencies itself. You just need to select an AI model, among other things, during setup, and you will be prompted through the entire process. Note that during the initial setup, you can select any AI provider you want. It does not have to be the same one you eventually use for coding. You can configure that later.
Once you are done with the initial setup, you need to replace the default interface with Discord. Start by creating a dedicated private server on Discord, then create channels relevant to your workflow. For example, I have agents, features, and reviews. Now you need to create an agent for Discord. Discord already has built-in bot functionality, so go to the Discord Developer Portal and create a bot there.
Once you have created the bot, generate a bot token. This token is essentially like an API key, so save it somewhere secure. Then invite the bot to your Discord server, select the required scopes, and give it permissions to view channels, read messages, and send messages. I would also enable reactions and messaging in threads to help keep conversations clean and organized. Now get your Discord ID and give the Discord token to OpenClaw. Create a config file to enable Discord in OpenClaw, restart OpenClaw, then return to Discord. Your agent should now be online.
Claude Code it is for me
Once your Discord interface is working, the next step is to connect the tools that will actually do the coding. OpenClaw should act as the orchestrator here, not necessarily as the tool that writes every line of code itself. I use Claude Code as the primary coding agent, but you could use Codex, OpenCode, Gemini CLI, or another supported coding harness instead. OpenClaw supports running external coding agents, such as Claude Code, via its ACP integration, so you can start and manage coding sessions while continuing to interact with the agent in Discord.
For my setup, Claude Code is the part that actually works on the codebase. It can inspect the repository, edit files, run commands, execute tests, and work with the rest of the development environment. I would give it access to a local clone of the project rather than expecting it to do all the coding through a GitHub integration. This gives the agent a proper development loop in which it can inspect the code, make a change, run the application or test suite, analyze the failure, and continue iterating until the task is complete.
I would still connect GitHub as a separate tool because it gives the agent access to the wider context around the codebase. GitHub has an official MCP server that can read repositories and code, work with issues and pull requests, and interact with workflows. This means you can eventually send a message in Discord such as, “Look at issue #142, find the cause of the bug, fix it, run the tests, and create a pull request.” The coding agent operates directly on the repository, while the GitHub integration helps it understand and manage everything related to that work.
Don’t forget the MD files
Markdown files tell the agent how your project works
The next important step is adding Markdown files that tell the agent how your project works. This is what stops it from approaching every task, as it has never seen your codebase before. OpenClaw gives every agent a dedicated workspace and automatically uses workspace files such as AGENTS.md, SOUL.md, TOOLS.md, IDENTITY.md, and USER.md as part of its context when they are present. You do not need to fill every file with pages of instructions, but a few well-written files make a huge difference.
I would start with AGENTS.md. This should explain what the agent is responsible for and how it is expected to work. I would also create a project-specific file that explains the application itself. You could call it PROJECT.md and include the architecture, important directories, current priorities, known technical debt, and any areas the agent should avoid touching. OpenClaw does not automatically inject every arbitrary Markdown file into context, so I would reference PROJECT.md explicitly from AGENTS.md and instruct the agent to read it before substantial work.
If you are using Claude Code, I would also add a CLAUDE.md file to the repository. Claude Code uses CLAUDE.md files to store persistent project instructions, including coding standards, workflows, architectural notes, and commands it should know about. This is separate from OpenClaw’s own workspace files.
You don’t need editors anymore
Well, that’s a bit of an exaggeration, but for most use cases, you don’t need an editor anymore unless you actually want to code things by hand. You can basically have your agent do things for you, and I think this is the direction editors are moving in as well. For example, VS Code now has an agent mode that doesn’t look like an editor at all. It just looks like an agentic chat interface. Cursor has something similar, and I’m sure this kind of setup will become much more mainstream very soon.
