Anthropic puts Claude Code inside Slack
Anthropic just integrated its Claude Code assistant directly into Slack as a research preview. Instead of copying and pasting bug reports into a separate tool, developers can now simply @Claude in a chat thread to pull context and linked repositories for quick debugging. You'll just need to connect your repos via the web version first.

If you write code for a living, you know the drill: your team diagnoses a nasty bug inside a chaotic Slack thread, and then someone has to manually copy-paste the whole mess into an IDE or a terminal to actually fix it.
Anthropic is trying to eliminate that alt-tab tax. The AI lab just launched an integration bringing Claude Code directly into Slack, available now in beta as a research preview.
From chat thread to open pull request
Previously, tagging Claude inside Slack only gave you lightweight chatbot assistance, like drafting a quick function or explaining an error message.
With this update, you can simply tag @Claude on a thread where a bug or feature request is being discussed. Claude will scan the conversation, recognize the task, and pull the thread's context automatically.
It then connects to the repositories you already authenticated in the web version of Claude Code. Claude figures out which repo is relevant, runs the coding session, drops progress updates directly into the Slack thread, and posts a review link or opens a pull request when it finishes.
You do not need to install a new app if your workspace already uses Claude, though you do have to configure your repos in Claude Code web beforehand.
The coding war leaves the IDE
This release shows that the battleground for AI engineering tools is rapidly moving out of code editors and into team communication hubs.
Competitors are already circling the same territory. Cursor offers Slack integrations for drafting code, GitHub Copilot can turn chat discussions into pull requests, and developers have wired OpenAI's Codex into custom Slack bots for months.
The launch comes just weeks after Anthropic released Claude Opus 4.5, which the company claims edges out Google Gemini 3 in coding benchmarks. Yet security questions linger; early tests revealed that Opus 4.5 refused malicious code generation requests only 78 percent of the time.
Handing Slack bot credentials access to internal code repositories also introduces fresh governance headaches and third-party dependencies. If Slack or Claude hits an API rate limit or goes down, your local engineering workflow can stall.
Why it matters
This shift signals that winning the AI coding market is no longer just about who has the smartest foundation model, but who captures the daily workflow of developer teams.
For engineers, the practical value is obvious. It drastically lowers friction for routine bug squashing and minor feature work by letting AI handle the mechanical handoff from product conversation to actual code.
We think developer teams already paying for Slack and Claude Code should definitely test this on low-risk repos. It saves time immediately, but engineering leads will need to think hard about repository access controls before unleashing it across company-wide channels.
Anthropic's play is clever because Slack is already the central nervous system for engineering teams. By anchoring Claude Code right where work is discussed, Anthropic creates serious pressure on Cursor and GitHub Copilot to match this level of ambient workflow automation.
Sources (2)
- theverge.com(opens in a new tab)Originaltheverge.com
- techcrunch.com(opens in a new tab)Reporttechcrunch.com