Weavable

Give every AI agent persistent work context

SaaS

Artificial Intelligence

Operations

Visit WebsiteSee on Product Hunt

The product looks like a genuinely useful tool, but it was shared to me by somebody on LinkedIn selling upvotes as a service pretending that it is their product.

Persistent context across agents is the exact problem most multi-agent systems hit. Workspace isolation makes context sharing safer — without it, cross-tenant leakage is a real risk. How are you handling tenant boundaries?

Really interesting launch, Abesh 👏 The continuous changelog approach feels like a big step forward compared to static RAG or raw API feeds. How do you see teams balancing the flexibility of shared contexts with the need to keep permissions tightly scoped as they scale?

Congrats on the launch. The activity graph / changelog framing is strong.

How do you decide which context should become a reusable workflow signal versus just being retrieved for one agent query?

This is very cool guys, congratulations. How does Weavable do in terms of speed at a practical level compared to connecting say claude code into all the individual data sources?

We've found connecting to the data sources directly is slow as well as being token heavy, claude has to pull some data, build a context, then pull more data, from that figure out what else it needs, it goes on for a while, especially if some are through MCP, does this help with that as well?

Yes! I've been trying to solve this problem for months with various (often questionable) hacks. Love it.

One thing I’ve been thinking about a lot with agentic systems is context governance.

Most team have hugely different sensitivity levels across data, customer conversations, board discussions, HR issues, commercial terms, etc. How does Weavable handle permissions and context boundaries so agents only reason from the information that specific users or teams should actually be able to see?

This is a really interesting point of view here. The activity-graph approach makes sense — context should reflect what's happening, not just what was recorded.

One question from our experience building Faindo: we connect to multiple AI models (ChatGPT, Perplexity, Gemini) and one challenge we keep hitting is that each model interprets the same context differently depending on how it was trained. Does Weavable normalize context before it hits the MCP endpoint, or does it stay model-agnostic and let the agent handle interpretation?

Congrats on the launch, following the progress closely.

Massive congrats on the launch! 🚀 One-tenth the tokens vs direct app connections, with 85% preference in LLM-as-judge evals — that's a serious pair of numbers to lead with, and it maps to a real pain. Most agent setups I've seen either drown in raw API output or reason from a snapshot that's already wrong. Treating context as live infra rather than a dump or a freeze is the right call. Signing up.

Nice! I especially like the activity graph/changelog approach because it treats context as something dynamic.

Curious: how do you actually reduce token usage by 90%?

Congratulations. And happy product launch. @abesh_thakur

Jumping in as the other maker.

Here’s the bet underneath everything we built: work isn’t just documents or records. It’s activity. The things people and agents do over time. A renewal slips because three signals lined up across CRM, support, and Slack that nobody connected. A deal closes because of a conversation in a thread, not a field.

The record is the residue. The work is what moved.

Most AI context tools either flatten all of that into a snapshot, or stitch together a handful of MCPs that make endless calls against flat records, pollute the context window, and still don’t know what changed or why. We thought both were wrong.

So we built Weavable on a deterministic engine that tracks how information changes, builds a changelog of every meaningful update, and stitches it into an activity graph. That graph is what your agent queries through the MCP endpoint. Not a summary, not a vector blob. A structured, time-aware picture of what’s actually happening. And because your agent can query for the specific signals it needs, it doesn’t ingest an entire workspace to find them. Less context window, less cost, sharper answers.

Would love to hear from anyone who’s tried to solve this differently. We think the activity-graph approach is the right primitive, but we’re early enough that we want to be wrong out loud if we are.