Braido is where separate threads — models, services, judgment calls — are plaited into a single workflow that holds its shape run after run. No code, no subscription; each pass through the loom is metered in power tokens.
1 token = $0.0002
Every part of Braido exists to take separate threads and return one finished piece.
Agents are your threads: a researcher, a stylist, a fact-checker — each with its own role, temperament and model.
Tasks interleave on a visual canvas. One agent's output becomes the next one's material, in the exact order you plait them.
120+ connectors feed the loom — RSS and webhooks bring material in; Slack, Telegram and HTTP carry the finished work out.
Bring your own Anthropic, OpenAI, Gemini or xAI keys and match each thread to the model it deserves. Keys rest AES-encrypted.
Each run is entered into an append-only token ledger, debit and balance-after together — the workshop's permanent record book.
A run that fails releases its token reserve entirely. You pay for finished cloth, never for a dropped stitch.
Watch three agents turn raw feeds into a publishable issue — and leave one honest line in the ledger.
Select the week's strongest threads
Spin them into publishable prose
Check the weave, tie off the issue
“Our weekly digest was four people and a Sunday evening. Braido plaits it from the same sources in eleven minutes, and the editor thread still catches what matters.”
“I treat the ledger like a record book. Every run, one entry, balance after — my accountant asked what software it was.”
“The zigzag from research to draft to review used to happen in my inbox. Now it happens on a canvas I can actually see.”
“We snapped a thread mid-run during a provider outage. The reserve came back on its own — that single detail bought my trust.”
“Bringing our own Anthropic and Gemini keys meant no new vendor negotiation. We picked the model per thread and moved on.”
“Braido is the first automation tool my writers did not resist. It reads like a craft bench, not a control panel.”
Buy a pack of power tokens; the loom draws them down as it works. Larger packs cost less per token, and nothing about it recurs.
| pack | price | tokens | ≈ runs | includes | |
|---|---|---|---|---|---|
| Starter | $9 | 25,000 tokens | ~56 | All 6 model providers · Sequential process · Built-in tools & actions | Top up |
| Makerpopular | $29 | 100,000 tokens +10,000 bonus tokens | ~244 | Hierarchical orchestration · Public API + 3 keys · Crew memory & templates | Top up |
| Scale | $119 | 500,000 tokens +75,000 bonus tokens | ~1,278 | Unlimited API keys · Priority queue lane · Low-balance alerts | Top up |
1 token = $0.0002 · Every pack is a one-time purchase. Balances carry forward indefinitely.
Never. Braido has no plans and no renewals — you purchase power tokens when you wish, and runs draw them from your balance. The ledger records each movement permanently.
Model usage and connector activity convert into whole power tokens. The run's debit is entered in an append-only ledger together with your balance after it, so cost is never an estimate.
Tokens are reserved before the loom starts and the reserve is released if the run fails. Incomplete work is never billed.
Yours, if you like: bring keys for Anthropic, OpenAI, Gemini, xAI or any OpenAI-compatible endpoint and set the model thread by thread. Keys are AES-encrypted at rest.
Yes — every crew is callable over a REST API with bearer keys. Keys are stored hashed, shown in full only once, and can be revoked instantly.
No. The shelf holds 330+ finished patterns — editorial desks, research digests, monitors, triage flows. Clone one and adjust the threads to your material.
An account takes a minute, starter tokens are included, and the first braid can come off the loom today.
Open the workshop