Urgent.News

What's breaking now, across thousands of outlets.

Editions

AI

Slack Code taps into collective vibe, puts AI agents into the group chat

Developers can now invite the whole team along for their quality time with the coding bot

Slack Code taps into collective vibe, puts AI agents into the group chat

Slack has introduced Slack Code, an AI coding agent feature that puts coding agents in group chats for everyone to see. This multiplayer AI allows coding bots to work alongside developers in shared project channels, rather than solely in developers' terminals. Users can tag an agent from a conversation, create a code channel for the job, and monitor its progress, review proposed changes, view live HTML previews, and guide its actions when necessary.

Once completed, the channel can archive itself, creating a searchable record of the process. Slack aims to involve more than just developers, as product managers, designers, and other staff can participate by spotting bugs, asking agents for fixes, and reviewing resulting diffs. The company emphasizes human oversight, with the ability to pause, redirect, or stop agents and approve code pushes to production.

Slack's code channels are compatible with various AI agents from companies like Anthropic, Cognition, GitHub, OpenAI's ChatGPT, and Vercel, with the APIs planned for broader developer access. Additional features include Agent DMs for direct conversations, an Agents tab for managing agents, and an Add to Slack feature for deploying third-party agents.

Slack positions Slack Code as a way to make AI-assisted coding more visible and collaborative, potentially revolutionizing the development process.

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

Read the original at theregister.com →

More in AI

acc vs acc_norm: Why Length Bias Skews LLM Eval Scores

Your fine-tune gains three points of acc_norm on HellaSwag and loses two points of acc . Same checkpoint, same harness, same seed.

  • "acc" measures highest summed log-likelihood of candidate continuation
  • "accnorm" divides sum by byte length of continuation string
  • Length bias can skew LLM evaluation scores significantly

More from Thursday 20 August →