Slack is launching collaborative vibe-coding channels
Slack has launched "Slack Code," a feature that gives teams dedicated channels for collaborative "vibe-coding" with AI agents, aiming to keep the whole process within Slack rather than requiring users to switch between separate tools and conversations. When a task such as building a feature, updating a page or fixing a bug arises, users tag in a coding agent (such as Anthropic's Claude or Cognition's Devin), which then spins up a code channel where the wider team can view the discussion, review code changes, preview outputs and approve the work before it goes live.
Each code channel automatically archives itself once its task is complete and keeps an audit log for later reference, part of Slack's push to frame AI agents as collaborative "teammates" rather than standalone tools. Slack Code is available immediately on all Slack plans and works with agents from its marketplace, with founding partners including Claude Code, Devin, Vercel Agent and GitHub Copilot said to integrate directly with the new channels.
- Slack launches "Slack Code" channels for team AI coding collaboration
- Users tag agents like Claude or Devin to handle tasks
- Available now on all Slack plans, integrates with major coding agents
New here? Start with this
Slack is a messaging service that many workplaces use to organise conversations into channels for different teams or projects. AI coding agents are software tools that can help write or change computer code in response to instructions.
In a code channel, a team can follow an agent’s work, discuss proposed changes and review previews before the changes are put into use. Slack’s approach brings those steps into the team’s existing workspace, where colleagues can see the discussion and refer back to a record of the task.
The feature connects Slack with coding agents made by companies including Anthropic, Cognition, Vercel and GitHub. It reflects a wider shift towards using AI tools in software development, while leaving teams to decide how they review and approve the work.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Slack Code eliminates context-switching by keeping development conversations unified in one place, where teams can discuss, review and approve AI-generated code without switching between tools. This increases transparency, ensures all stakeholders stay informed, and treats AI as a true collaborator rather than an isolated tool—reflecting how teams actually work and reducing workflow friction.
The case against
Professional code development requires purpose-built systems with robust security, granular access controls and audit trails that communication platforms cannot replicate. Code review in Slack risks degrading engineering discipline and encouraging casual approval without rigorous scrutiny, whilst development still fragments across multiple systems (Slack, repositories, IDEs), and dilutes the structured processes that maintain code quality, security and compliance.