Urgent.News

One page, thousands of outlets. See who else covered it.

Editions

AI

Slack launches Slack Code, adding dedicated, project-specific channels that let teams collaborate with AI coding agents "like teammates" across all Slack plans (Jess Weatherbed/The Verge)

Slack Code provides a dedicated open space for teams to collaborate with AI agents.

Slack has introduced Slack Code, a new feature that allows teams to collaborate with AI coding agents in dedicated, project-specific channels. Available on all Slack plans, this feature aims to streamline collaboration by providing a single space for coding tasks, rather than switching between various tools and conversations.

Slack Code enables teams to work alongside coding agents such as Anthropic's Claude or Cognition's Devin within these dedicated channels. These agents can create a code channel to handle tasks like building features, updating web pages, or fixing bugs. The channel provides real-time previews of the agent's output, live feedback, and an audit log for record-keeping. Once the task is completed, the channel automatically archives itself.

Slack is positioning this feature as a way for teams to collaborate with AI agents "like teammates". The integration will work with agents available on Slack's marketplace, including Claude Code, Devin, Vercel Agent, and GitHub Copilot, which are from Slack's founding partners.

Written by urgent.news from Techmeme's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at theverge.com →

More in AI

A benchmark is only as good as the model you use to grade it

I built a pytest harness that runs the same set of questions through five language models at once - a free local Llama, plus GPT, DeepSeek, and two Claude models - and compares them on the three…

  • Benchmark worth depends on grading model
  • Paid models scored equally in quality
  • Llama underperformed due to leniency in scoring

More from Thursday 20 August →