API Terms

Effective July 6, 2026

These API Terms govern your use of the Jinn gateway - the Model Context Protocol (MCP) API at app.jinn.works/api/mcp - and any token issued to you (Connected or Agency). They supplement, and are part of, the Terms of Service. By connecting an agent with a Jinn token, you agree to them.

1. What the API does

The gateway is a read-onlyendpoint that serves your own brand’s Brand DNA - identity, voice, positioning, messaging, and (on paid tiers) your product, competitive, and design data - to your AI agents in machine-readable form. It does not write to your systems and it exposes no other customer’s data.

2. Your token

Access is by bearer token. Every token is read-only and scoped to the brands on its allowlist - it can never read a brand you don’t own. Treat a token like a production API key: it is shown once at creation, and you are responsible for all activity under it. Store it in a secret manager, never paste it back into an agent chat, and rotate or revoke it from your token panel at any time. We may require rotation after a suspected exposure.

3. Acceptable use

You agree not to:

  • Access, or attempt to access, any brand or data outside your token’s allowlist
  • Probe for, enumerate, or map brands, tools, or data you are not entitled to
  • Reverse-engineer, scrape, or attempt to reconstruct our models, extraction pipeline, or data schema
  • Exceed or circumvent rate limits, or otherwise interfere with or degrade the API
  • Share one token across parties as a way to avoid per-brand or per-seat terms
  • Use the API for any unlawful purpose

4. Rate limits

The API is rate-limited per token (currently 120 requests per minute). We may adjust limits, throttle, or apply usage ceilings to protect the service; ceilings sit well above normal use, and we alert rather than silently cap wherever we can. Automated outlier monitoring may flag anomalous volume on a token.

5. No resale of projections

The Brand DNA projections served over the API are licensed to you for grounding your own brand’s agents. You may not resell, redistribute, sublicense, syndicate, or otherwise make them available to third parties, and you may not use them to build or train a competing dataset, model, or service. Your own brand data remains yours; this restriction is on the machine-readable projection and the methodology it encodes.

6. Subscriptions & billing

Connected and Agency are monthly subscriptions billed in advance through Stripe and renewing automatically until you cancel. A Brand DNA report purchase includes a 30-day Connected trial; the trial clock starts when your token is issued. You can cancel any time from your billing settings - cancellation stops future renewals and takes effect at the end of the current period. Fees already paid are non-refundable except where required by law. Founding pricing, where offered, is honored for as long as your subscription stays active; published list pricing applies to new subscriptions once a founding window closes.

7. Suspension & revocation

We may suspend or revoke a token, or a brand’s access, immediately for breach of these terms, abuse, security risk, or non-payment. On cancellation or refund, your brand’s tokens are revoked and its projection stops being served; your extracted Brand DNA is retained but no longer readable over the API. You can revoke your own tokens at any time from the panel.

8. Disclaimers & liability

The API is provided “as is.” Brand DNA is generated by automated analysis and may be incomplete or inaccurate - you are responsible for reviewing what your agents produce with it. To the maximum extent permitted by law, our total liability is limited as set out in the Terms of Service.

9. Changes & contact

We may update these API Terms as the product evolves; material changes are posted here. Questions? Email dan@jinn.works.