Site navigation

For AI-native founders

Stop babysitting AI that doesn't know your company.

An AI team that knows your company and runs its work, with receipts.

One task in. One receipt out.
Reads
  • Slack
  • GitHub
  • Notion
  • Linear
  • Google Meet
Seats
  • Claude
  • Codex
  • Cursor
  • Grok

open the API to customers

M-12

needs you 1verified by Grokc7a91e2

API access is included in Pro

paid add-on · scrapped 14 Jan

Wave 1
  • Issue and verify API keysPR #251Claudecappedcontinued on Codex at 09:216m 41s6m 41s
  • Rate-limit by plan at the edgePR #251Cursordone
Wave 2
  • Publish the API referencedocs/apiOpenCodedone

Verdict

Grok

a different family

  • Cookie sessions unaffected
  • Limits enforced before the handler
  • Rate-limit headers on 429
  • Keys never stored raw
  • Docs match the routes

Rate-limit headers on 429

Codex · fixed
PR #251held for your merge~$3$2.90quoted ~$3, settled $2.90
KNOWSRUNSRECEIPTS

You'll explain your company

ClaudeNew chat

What's the status of the public API launch?

I don't have visibility into your projects. Could you share where things stand?

Reply to Claude…

Cursorv0/routes.ts

Add the v0 deprecation notice.

I don't see a launch date in this repo. What should it say?

Plan, search, build anything

ChatGPTNew chat

Draft the launch email to customers.

Happy to. What are you launching, and when?

Ask anything

Eventually you stop checking.

KNOWSRUNSRECEIPTS

Close the tabs.
Keep the decisions.

Forward the thread. Review the change once. The next session starts from the corrected file.

#launch

Ship the public API. Don't slip.

API launch research

Keep v0 working through the launch — dual-run until the new routes are in.

PR #412

Sunset notice merged.

ENG-241

Must clear before anything ships.

launch/public-api.md

Public API launch

Launch date
March 10March 17
Migration path
Keep v0 working through the launch
Sunset notice
Merged
Security review
Must clear before anything ships

launch/public-api.mdWhat should the API announcement say?

Lead with March 17 and the v0 migration path — security review clears first.

KNOWSRUNSRECEIPTS

Hand it a task. It runs on the seats you already pay for.

One job in, whichever seats it needs, and it keeps going when one runs out.

Four AI seats plugged into the top of one machined deck, with a task riding the conveyor across its front. The seat working that task is lit; the others sit dark.
  • Claude Max · GA · Mac
  • Codex · preview
  • Cursor · preview
  • Grok · preview
  • One open slot
  • Pull request, held for your merge
  • Draft, sourced
  • Held for your send
KNOWSRUNSRECEIPTS

Watch one run, start to finish.

One task, two touches from you: approve the plan, read the review. Everything between them runs without you.

01

It starts fully briefed

You type five words. What your company already settled comes with them — and the one thing it never decided, it will not guess.

TPublic API v1T-241
Plan
Running
Ready
JB

You

open the API to customers

Brief

  • Three customers are waiting on this
  • API access is included in Proreplaced the paid add-on, scrapped 14 Jandecided 4 Feb
  • Keys rotate every 90 daysproposed in #eng, never decidedleft for you

Claude

Reading your codebase61 files
  • Readingmiddleware/auth.ts142 lines
  • GrepAuthorization0 matches
  • Readingpackage.jsonno rate-limit dep
02

It splits the work across your seats

Claude, Codex and Cursor each take a piece, signed in on your Mac. When one hits its cap, the next one continues on Codex, briefed.

TPublic API v1T-241
Plan
Running
Ready
Two ways in, one session objectapproved by you09:14

Wave 1· at the same time

  1. ClaudeIssue and verify API keysdone · 6m 41s
    cappedCodexcontinuing 09:21
  2. CursorRate-limit by plan at the edgerunning

Wave 2

  1. OpenCodePublish the API referencequeued
03

A different family grades it

Five criteria, and a grader that did not do the work. One of them fails before it passes — and the fix is already in the commit.

TPublic API v1T-241
Plan
Running
Ready

Grok

a different family

2 need youreceipt #9c2f
  • Cookie sessions unaffected
  • Limits enforced before the handler
  • Rate-limit header missingRate-limit headers on 429Codex · fixed
  • Keys never stored raw
  • Docs match the routes

Your API is live on Pro. Nobody's session broke.

Worth flagging

  • Nothing expires a key yet

    Hashed at rest, shown once — but nothing rotates them. Your policy call.

  • Your sales deck still promises 600 on Free

    The run used decisions/rate-limits.md. The deck disagrees; I left it alone.

curl https://api.acme.dev/v1/usage -H "Authorization: Bearer sk_live_…"

What changed

SurfaceChange
api_keysnew table · hash, prefix, last used
POST /v1/keysnew · issued once, shown once
auth middlewarecookie or key, one session object
docs/apigenerated from the routes

Evidence

  • Cookie sessions unaffected47 existing tests green
  • Limits enforced before the handler429 at request 601
  • Rate-limit headers on 429Retry-After on every reject
  • Keys never stored rawhash + prefix only
  • Docs match the routesgenerated · 0 drift
Undo38 min · $2.90 of ~$3 · c7a91e2

You touched it twice.

KNOWSRUNSRECEIPTS

Why can I stop watching?

It finishes every task it can verify and undo. Anything it can't stops and asks you — that's the only queue you watch.

It finishes every task it can verify and undo. Anything it can't stops and asks you — that's the only queue you watch.

General taskNeeds youRenewal price for Northwind2:10 PM

Supervisor recommends $499/mothe list price in enterprise-gtm.md. Customer-facing and can't be undone, so it stopped.

Renewal price for Northwind
KNOWSRUNSRECEIPTS

Run your company, not your tools.

Headcount of one. Leverage of a team.

Built for you ifyou pay for 2+ AI subscriptionsyour company lives in 4+ toolsyou're the one carrying context between them

Keep the tools you already use.

  • Claude
  • Cursor
  • Codex
  • ChatGPT
  • Grok

They read your company from one place.

Add it to Claude Code or Cursor

Or hand it the task.

  • Slack
  • GitHub
  • Notion
  • Linear
  • Google Meet

Free plan, no card. Nothing acts until you turn it on.

Frequently asked questions

What do I actually get on day one?

Connect your sources, or bring in a research thread. Get a cited company record and use it in your next coding or content task. You approve work before it runs. AI Squads are a preview of what comes next, not a requirement to get started.

I already use Cursor and Claude Code — what does this add?

HiveBase works with the agents you already run. It briefs them from your company, routes work across the models and keys you already pay for, and brings back a receipt. Cursor stays Cursor; Claude Code stays Claude Code.

How is this different from a bot signed into my tools?

HiveBase connects the decisions in your chats and documents to the work your agents do. A task starts with company context and returns with evidence of what changed. You can inspect the sources and decide what is allowed to proceed.

What does it replace?

The job, not the tools. Your Cursor stays Cursor; your Linear stays Linear.

Do I keep my CLAUDE.md / GBrain / Cursor rules?

Yes — overlay. CLAUDE.md, GBrain, and Cursor rules stay yours. HiveBase keeps them true. Delete us and you keep everything.

What does it cost?

Reads are free. Work you ask for comes out of your plan. Nothing meters what it merely watches. Runs on the models and keys you already pay for, no markup on those.