Salesforce launched Slack Code on August 20, making AI-assisted software development a shared team activity rather than a private exchange between one developer and one agent. The feature embeds AI coding agents directly into dedicated Slack channels where the whole team can watch, steer, review, and approve work before anything ships.

It is a small but meaningful shift in how businesses think about AI in their software workflows. Instead of one person quietly prompting an agent in a terminal and committing the output, Slack Code puts the entire conversation on the floor for the team to see.

How It Works

A team member mentions a supported agent into a project channel. The agent then opens a dedicated code channel for that session. From there, every team member can see the same conversation the agent is working from, inspect proposed changes as code diffs, preview HTML output live, leave feedback the agent incorporates, and sign off before anything gets merged.

When the project wraps up, the channel can be automatically archived, and audit logs keep a record of every action the agent took. For high-stakes steps like pushing to production, the workflow requires a human approval before proceeding.

The supported agents at launch include Anthropic’s Claude Code, Cognition’s Devin, GitHub Copilot, ChatGPT, and Vercel’s agent. Slack Code is available across all Slack plans.

Why This Matters for Businesses

The productivity gains from AI coding agents are real. The problem most teams run into is auditability. When one developer owns the entire agent interaction, the rest of the team has no visibility into what got built, why, or what the agent decided to change. If something breaks in production, it is hard to reconstruct what happened.

Slack Code shifts that dynamic. The coding channel becomes the record of truth. Product managers can see what the agent is building and flag issues before code ships. Security and compliance teams can review the audit log. Junior engineers can watch how the agent approaches a problem and learn from it. Everyone is working from the same shared context.

For enterprise teams already running AI agents across other workflows, this is a logical extension. The same governance controls that IT uses for the rest of Slack, including permissions, admin controls, and data policies, apply to Slack Code without additional configuration.

What This Means for Business

If your business is exploring AI coding agents, Slack Code addresses the two biggest objections that tend to slow adoption in team settings: lack of visibility and lack of control.

Most organisations that have experimented with AI coding tools report that getting the agent to write working code is not the hard part. Getting the rest of the team comfortable with AI-generated code shipping to production is. Human approval gates and shared channel context go a long way toward solving that.

The agent lineup at launch is also telling. Anthropic’s Claude Code, Devin, GitHub Copilot, and Vercel’s agent represent the tools that development teams are already using independently. Bringing them into a shared Slack channel reduces fragmentation without requiring teams to pick a single vendor.

Broader agent APIs for other business functions, including marketing and legal review, are on Salesforce’s roadmap but not yet available in this release.

The Bigger Picture

Slack Code is part of a broader pattern playing out across enterprise software this year: AI agents are moving from individual productivity tools into team-layer infrastructure. The question is no longer whether an AI agent can do the task. It is whether the team can trust, verify, and govern what the agent did.

Shared channels with audit trails and approval gates are one answer to that question. Businesses building out their AI agent strategy should take note: governance and visibility are becoming the competitive differentiator, not just capability.

If you are thinking about how AI agents fit into your operations, that is exactly the kind of strategic question our Omni Advisory service is designed to help with. Talk to the team about building a governed, scalable approach to AI in your business.