The Problem: AI Content Still Requires Too Much Manual Work
Every content team has the same story. You open ChatGPT, write a prompt, and get a draft in seconds. Then the real work starts.
You fact-check the claims. You rewrite for brand voice. You send it to a subject-matter expert. They send it back with corrections. You re-verify the changed sections. Two to four hours later, you have something publishable — for one article.
Multiply that by 10 articles a week across 3 client brands, and you've got a full-time job that's 80% cleanup.
The draft was never the bottleneck. The pipeline after the draft is.
What's Now Possible: The J77.ai API
Today we're opening the same multi-model pipeline that powers J77.ai as a REST API. Everything you can do in the web app — create brand kits, start builds, generate images, manage your content library — is now available programmatically.
Here's what that looks like in practice:
# Create a build
curl -X POST https://api.j77.ai/api/v1/builds \
-H "Authorization: Bearer j77_live_YOUR_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: weekly-blog-2026-w08-post-1" \
-d '{
"targetId": "your-target-uuid",
"buildType": "single",
"objective": "Write about reducing SaaS churn with proactive customer success",
"contentKind": "blog_post",
"researchEnabled": true
}'
That single request kicks off the full J77 pipeline: real-time web research, multi-model drafting, critique, claim verification, and brand alignment. When it's done, you download publish-ready content with sources attached.
The Use Case: Batch Content Overnight
The most powerful pattern is batch automation. Here's the workflow agencies and SaaS teams are already building:
- Evening: A cron job creates 20 builds — one per article in your editorial calendar. Each build uses an idempotency key, so retries are safe.
- Overnight: J77's pipeline researches, drafts, critiques, and verifies each piece. No human in the loop.
- Morning: Your team downloads 20 publish-ready articles with sources, claim verification, and brand voice already applied. Review takes minutes, not hours.
The math changes completely. Instead of 2–4 hours of cleanup per article, you're spending 10–15 minutes reviewing content that's already been fact-checked and brand-aligned.
What You Get
- All v1 endpoints — targets, builds, tokens, library, images, chat, suggest
- Scoped API keys — create keys with
read,write, andbuildspermissions. Rotate and revoke anytime. - 60 requests/minute — with dedicated limits for builds (10/min) and media endpoints (20/min)
- Idempotency — send an
Idempotency-Keyheader with build requests to prevent duplicates - Same token system — your API and web app share the same token balance. No separate billing.
- Interactive docs — full Swagger documentation at
/api/v1/docs
Pricing
API access is $300/year. This unlocks programmatic access to all endpoints. Tokens are purchased separately through the same bundles available in the web app — starting at $0.50/token for the 100-token bundle.
Think of it this way: API access is the gateway, tokens are the fuel.
Getting Started
- Sign up or log in at j77.ai
- Subscribe to API access in your account settings
- Create an API key with the scopes you need
- Make your first request
Full documentation, code examples, and best practices are available on our Developers page.
The J77.ai API is available today. If you're an agency managing multiple brands, a SaaS team with content needs, or a developer building content workflows — this is for you.