About prxhub

Research should accumulate, not evaporate.

prxhub is the registry AI research never had. An open, citable home for verified research bundles, so the next person asking your question doesn't have to run it again.

Why we built this

AI research has a caching problem. Every day, millions of people ask AI systems the same questions. The same models run the same queries, return similar answers, and every single response is thrown away when the tab closes.

There's no shared memory. No registry. No "someone already figured this out." Just an endless churn of duplicated compute and evaporated knowledge.

We built prxhub to be the registry that AI research never had. A place where a well-run research bundle (multi-provider, sourced, attested) can be published once and used forever.

The problem, in depth

AI answers aren't citable.

Students cite Wikipedia. Researchers cite papers. Nobody can cite an AI conversation, because there's nothing to point to. The answer lives in one browser session and nowhere else. prxhub bundles are permanent URLs. Cite them like you'd cite a report.

Research disappears when the tab closes.

You spent 30 minutes iterating on a query, got a great synthesis, closed the tab. Next week you need it again. You start over. prxhub stores the full bundle (query, per-provider reports, synthesis, claims, sources) so nothing you build disappears.

Everybody re-runs the same work.

Right now, analysts across the world are asking similar questions to the same five AI providers and getting nearly identical answers. None of them know the others exist. Every query burns compute, water, and time. When you publish to prxhub, the next person searching that topic might not need to run anything. That compounds.

Our philosophy

The best knowledge systems in history are cumulative. Scientists cite prior work. Engineers reference specs. Developers clone repos. Knowledge compounds because it's preserved.

AI research is the exception. It's almost entirely ephemeral. We think that's a solvable problem.

prxhub is our attempt: an open registry where AI research bundles are first-class artifacts, just like code is on GitHub or models are on Hugging Face. We're not trying to replace your AI tools. We're building the layer that makes their output last.

Who it's for

Researchers

Publish attested findings with citable sources. Stop re-running what's already been answered.

Analysts

Multi-provider synthesis with claims and confidence scores. Build on your team's prior work.

Developers

.prx is portable. Bring your research into any pipeline. MCP-native, API-first.

Teams

Share a research collection, fork a colleague's bundle, build a living knowledge base together.

The environmental and ethical angle

We don't lead with this, but it matters. AI systems use a lot of compute. Compute uses energy and water. Most of that usage is repetitive: the same questions asked by different people, answered fresh each time.

When research bundles are reusable, that duplication shrinks. We can't fix AI's footprint, but we can reduce the part caused by redundancy.

On attribution: AI aggregates. That's not inherently bad, but it can obscure where ideas come from. prxhub bundles store sources per-claim. Every synthesis links back to the underlying material. That's not perfect provenance, but it's better than a chat window that credits nobody.

For AI agents · use prxhub right now

prxhub is agent-native. Before your agent runs a new research query, it should check if the answer already exists. prxhub has a public MCP endpoint and a llms.txt. No account required for public read access.

Recommended workflow

Search existing bundles first. search_bundles via MCP, or GET /api/search/bundles?q=...
Pull the synthesis. If a bundle matches, download the manifest and read synthesis/report.md and synthesis/claims.json for machine-readable claims with confidence scores.
Only run new research on what's not covered. Use Parallect or your preferred research tool. Export as .prx, publish back to prxhub.

MCP config

{
  "mcpServers": {
    "prxhub": { "url": "https://prxhub.com/api/mcp" }
  }
}

Available MCP tools

search_bundles(query, limit) finds relevant existing research
search_claims(query, limit, confidence?) finds specific claims with sources
download_bundle(slug) returns a presigned URL for the raw .prx archive

Install the CLI for deeper work

pip install prx-cli
prx search "your topic"
prx clone username/bundle-slug
prx read results.prx

No account. No API key. Start using prxhub in your agent workflows today. Full reference: prxhub.com/llms.txt.

The vision

We want prxhub to be the place where AI-era knowledge accumulates. Not a chatbot. Not a search engine. A registry.

The place you go before you ask the question, because someone probably already asked it, and published what they found.

If that vision appeals to you: publish your first bundle. Contribute to someone else's. Build a collection. The registry is only as good as what's in it.

Get started Explore existing bundles