AI agents are changing how we work and collaborate, but documents are still stuck in the pre-AI era. Agents can now create beautiful, interactive decks, dashboards, and reports in HTML. But the moment you want to collaborate on them, everything breaks down. Google Docs wasn’t built for rich, interactive... HTML or AI agents. Feedback gets scattered across Slack and email. Every edit means going back to the agent, generating a new file, and sharing another link. No real-time collaboration. No single source of truth. We built SlashDocs to fix this. Think Google Docs, but built for AI agents and the documents they create. Host HTML and Markdown in one central hub. Share documents, manage permissions, and comment directly on the content, just like Google Docs. Agents can read feedback and update the same document in real time. No duplicate files, no endless back-and-forth, and everyone stays in sync. What started as a small internal experiment in one team has now scaled across Razorpay. 28,000+ documents created. Over 50% of all documents generated at Razorpay now live on SlashDocs. And today, we’re opening it up to everyone. Self-hosted. Agent-agnostic. Fully open source. AI agents have changed how we create documents. It’s time we changed how we collaborate on them. Try it:show more

Congrats on open sourcing this. I tried self hosting it today and put together a Railway template with Postgres, S3 storage and a Dex login included, so the whole stack comes up in one go and you can sign in right away. Sharing in case it helps anyone who wants to try it without a local setup:

I use Cursor to make quick internal tools, but the handoff is always awkward. Never know where to host them & share with the team. This is super cool

I’ve sent raw HTML over Slack more times than I’d like to admit. This is a cool fix.

Love the fact that everything under razorpay umbrella is about being sharp or cutting something.

28k docs inside one company is a hell of a bake before open sourcing does the agent pull comments as a tool call, or does it watch the doc like a human would?

documents are fine AI agents are the problem

Cool stuff. TRYING NOW.

documents as first-class agent citizens is the move. treating them as static blobs that agents read breaks when collaboration requires live state, not snapshots.

does a comment stay attached to the right section after the agent rewrites it?

28,000+ internal docs before open-sourcing this is a pretty serious dogfood story.

downloading this tonight.

This is cool. Will add it to my personal harness this weekend to test. Very curious about the BI capabilities!

With those HTML decks, where does collaboration break first: comments, shared editing, or keeping everyone on the same version?

It can break at all 3 but SlashDocs solves for all and has become the default for us.

Having a default saves a discussion about where everyone should work on the next deck.

agents that read the comments and edit in place is the unlock. the comment thread is the real product

cool stuff

Documents are the last thing still emailed as attachments, and Google Docs predates the agents that now write them.

Google Docs already does real-time collab. The gap is agents sitting outside the loop.

Thank you. I build something similar at my company, but commenting and sharing is what i was looking

asking your colleague to open a pull request to move a pie chart is a tough sell

So true — docs are the last thing to catch up. Agents need living documents that update themselves, not static files that go stale after one sprint.

Agents can now produce interactive HTML decks and dashboards, yet collaborating on them is where the workflow falls apart. Would love to have you in our network, let's connect.

Real collaboration needs a single shared doc source.

Great share

Super

how do agents read the comments, mcp?

I’ve got a folder called new-dashboard-final2.html. Enough said.

I'm gonna have so much fun creating interactive docs

Great solution. If we can share this (HTML) interactive pages with clients it will replace PPTs & PDFs of product decks.

Self-hosted matters here. These are the kinds of docs I wouldn’t want sitting in a random third-party workspace.

How does this differ from claude artifacts? Especially if the org is on team plan?

Agents are making creation cheaper, so coordination is the next bottleneck. A shared, editable source of truth could become essential as more work happens between humans and agents.

Nice,but how its different from Confluence with its MCP server for storing agent generated docs in HTML?

imma start shipping our weekly project tracker and see if anyone notices it’s no longer maintained manually.

Great share

The next challenge is making sure collaboration produces lasting organizational knowledge, not just better documents. When agents revise the same document based on human feedback, can that feedback, the reasoning behind each change, and the final outcome become traceable and reusable across future workflows? A shared document is a great start. A shared, evidence-backed knowledge layer is the next step.