Your engineering team is about to snap. And your AI coding agent is making it worse. Introducing CodeRabbit Agent for Slack 🎉 A second brain for engineering teams. Because your tribal knowledge lives in Slack threads nobody can find. At CodeRabbit, we review millions of PRs every week and... know how ace engineering teams operate on the planet. The same three things slow every team we see: - Context and decisions that live elsewhere. - Lack of a team-level durable knowledge base - And a trust layer that gives a safety net to your teams Built for Agentic SDLC workflows, CodeRabbit Agent for Slack solves all three problems in one shot while enabling teams to collaborate in real-time with the Agent. CodeRabbit Agent is Slack-native and builds your team’s operating context from every thread, every decision, every conversation your team has ever had. Make your team’s context compound!show more

Try it for free! Get $50/user free agent minutes.

the wildest part: your "senior engineer" right now is whoever remembers why you deployed on Tuesdays. that's not seniority, that's a single point of failure

This is why onboarding takes 6 months at most companies. New engineers aren't slow learners, they're just waiting for the one person who remembers why things are the way they are to have a free 30 minutes. The tribal knowledge tax is massive.

Does it work with my printer?

This looks good building a shared memory layer where that context already exists makes a lot of sense. This could remove a surprising amount of repeated work across teams.

The repeated work problem is massively underrated. Every new engineer, every new agent session, every new Slack thread starts from zero. A shared memory layer compounds instead of resets.

The first instance I’ve seen of multiplayer AI

Time to try it out - we're live at

This is finally here, we have been cooking and using it internally for sometime already, give it a try

me following this release from Miami

looks great

Thanks, Michael. Please give it a try!

This is so real. Context switching is honestly a bigger problem than code complexity at this point.

with agents, we're now re-explaining the context and losing time & tokens ^ This is the solution to it

Every agent interaction today is re-priming context from zero. We're betting the next unlock is persistent team context the agent already has.

congrats Harjot, @aravindputrevu and team 👏🔥

The new Slack agent proly targets one of engineering’s oldest problems: knowledge that disappears into conversation.

Exactly what we've set out to resolve with

Very nice approach to handling a huge emergent AI issue. Looking forward to kicking its tires!

Would love for you to try out the product & share any feedback. Get $50/user free agent minutes:

Thanks

this fixes basically the biggest issues I’ve seen with Claude Code: 1. zero memory - every session starts from scratch 2. no team context - only knows what I tell it 3. decisions die in Slack threads no one can search living directly in Slack is huge.

Yes, that was the thesis. Code review context shouldn't live in one dev's head or die in a Slack thread. Slack is where the decisions actually happen, so that's where the agent needs to be.

context problem is big in workflows. congratulations @harjotsgill and team

Yesss this is awesome It’s the lack of knowledge and team context that has been making things harder. Looks great 🫡

Thanks @JamesonCamp! Would love for you try out the product and exchange notes:

no more re-explaining context to a bot that forgets everything at 5pm

Teams we worked with were burning the first 20 minutes of every session just re-priming the agent. Multiply that across 10 engineers and the math gets ugly fast.

the moat was never the model lol. it was context. took everyone way too long to figure this out

100%. Context is the only durable moat: who your team is, what you've shipped, why you made the calls you did. That's what we're building.

Lessgo man, count me in!

Harjot, I'm so impressed with what you guys have built. I remember the early days. It's nuts how far you've come. This update feels spot on. Many teams I talk to have moved to Slack as their primary surface for interacting with agents as a team, not coders, but everything else. This move captures that movement perfectly. Whats one of the biggest learning from using this internally?

We're shipping faster on Soft engg workflows: build failures, issue triage, RCAs. As the knowledge layer builds up, runbooks run themselves, and devs spend less time steering the agent.

We’ve recently had some insane conversations internally at @flocurve too, and parts of our system agents have become better and self-improve / optimise based on signals. It’s hard to say where this is going to be completely transparent

coderabbit agents catching bugs written by other agents;

Providing context saves time. Saving time saves money and allows people to work on more important stuff.

Let's go team 🚀

@Krishnasagrawal Very excited to test it

@Krishnasagrawal We're at

We’ve spent years optimizing how fast code gets written, but almost no time on how context is preserved. Every PR has a backstory. Every decision has a reason. When that disappears, teams slow down. This feels like fixing the actual bottleneck.

This is 🔥 Turning Slack chaos into structured, searchable context is exactly what teams need right now. CodeRabbit just made “tribal knowledge” actually usable 🚀

10 years behind greptile

Engineering context is the new competitive moat. Teams that systematize tribal knowledge through AI leverage unlock faster iteration. CodeRabbit solves the friction between intent and execution. This compounds over quarters.

good

good team

Congrats on the launch!!

Devs problem got solved

Wow amazing production!

Huge congrats, this is the best feature of CodeRabbit so far, let's go! 🚀

tribal knowledge locked in a Slack thread nobody can find is literally the bane of every dev team lol

yes!