Anthropic has updated its upcoming collaborative projects for Claude Code. During project creation, users can now add repositories directly as persistent context. The interface recommends adding only the repositories required for each session, and suggests attaching additional repos on a session-by-session basis when needed.
ANTHROPIC 🔥: Claude Tag is coming to Claude Desktop in the form of collaborative Projects for Claude Code.
— 🚨 AI News | TestingCatalog (@testingcatalog) August 16, 2026
That's essentially a built-in Slack 👀
> Earlier spotted Managed Projects feature got a slight upgrade, mentioning that users will be able to spawn different threads… pic.twitter.com/aOzCEK2m9Y
At its core, the feature is designed around shared memory and context that persist across sessions, with multiple people able to work inside the same session. The latest description also introduces multiple threads within a session.
A useful parallel is Slack: the project session effectively becomes a channel where several people and Claude are present, while threads allow separate sub-conversations around individual tasks. This fits Anthropic’s existing direction with Claude Tag on Slack, where Claude can use channel and thread context alongside connected codebases.

Anthropic has suggested this collaborative model could represent the future of software development. Instead of traditional pair programming, or a developer working alone with AI, multiple developers and Claude can work together on the same project. Anthropic has reportedly used this approach internally for some time, and managed projects could bring that workflow directly into Claude without requiring Slack.
Great research went into making Claude proactive. Our data is showing unprompted messages are down ~45% -- Claude Tag is both more context-aware across your work and more attuned on when to keep quiet.
— Noah Zweben (@noahzweben) August 13, 2026
Best part - you don't pay for Monitoring either!https://t.co/Tvo3347fY9
Anthropic's focus has been around Claude Tag recently
Another important piece we discovered earlier is that Claude is expected to maintain these projects itself. It can periodically revisit context files and memories to reorganize and refine them, turning the project into a persistent workspace rather than static shared chat history.
Availability also appears to be expanding beyond friends-and-family or alpha testing, with early-access organizations potentially already having access. There is no confirmed public rollout date yet.
This may look like a relatively small addition at first, but it could become one of Anthropic’s larger workflow releases. The company is increasingly positioning Claude as infrastructure for shared organizational work, rather than only a one-to-one coding assistant, and collaborative projects appear to be a direct extension of that strategy.