Slack has introduced Slack Code, a new feature that provides dedicated, project-specific channels where teams can collaborate with AI agents on coding tasks. The tool is designed to centralize the development process, allowing users to bring in coding agents directly within a shared space rather than switching between separate apps and conversations. Slack Code is available immediately on all Slack plans.
The feature works by letting users tag a coding agent, such as Anthropic’s Claude or Cognition’s Devin, directly into a Slack channel. When an agent is tagged to handle a task like building a feature, updating a web page, or fixing a bug, it automatically creates a new code channel for that assignment. All team members can then view the agent’s conversation, review code changes, and see live previews of the output before deciding whether to approve the work for release.
Slack Code includes several built-in tools to support this workflow. Users have access to dedicated tabs within the channel, a feature for comparing coding changes, and the ability to preview HTML output ahead of shipping. The channels are also designed to archive themselves automatically once the task is completed, and they include an audit log for record-keeping purposes.
Slack is positioning the offering as a way for teams to treat AI agents as teammates in the development process. The company is not alone in this approach, as the source article notes that SpaceXAI is similarly promoting its Grok Bot agents with a comparable mindset about collaborative AI work. Slack’s new code channels will work with the AI agents already available on its marketplace.
Founding partners for the launch include Claude Code, Devin, Vercel Agent, and GitHub Copilot, and Slack says these agents will integrate seamlessly with the new code channels. The feature is being rolled out across all Slack plans starting today, with no additional plan tiers or pricing mentioned in the announcement. The company frames this as a way to keep all aspects of a coding project visible and auditable within one platform.
More AI news from TechManNews.
