We store token counts and discard your prompts.

This applies to every request. Each response carries x-st-data-retention: none so your agent can check it.

Prompts and completionsNot stored

Processed in memory to produce the response, then discarded. They aren't written to logs.

Batch filesAt most 30 days

Stored encrypted in your region until you delete them or 30 days after upload, whichever comes first.

Usage recordsLife of the account

Model, token counts, cost and timings, for billing. No content.

We don't use your prompts, completions or files to train models.

Closing an account

Close the account with DELETE /v1/account (a key with the keys or billing scope), or, in the US region, as an owner on the dashboard. Every key stops working and all running work (batches, videos) is cancelled at once, and the account's files and batch results are deleted right away. For 30 days the account itself (its keys and members) can be reopened, from the dashboard in the US region or by contacting support; after that its personal data (name, email, key names) is erased. A legal hold can require us to keep data longer. Unused purchased credit is refunded automatically once usage has settled, so a reopened account keeps only credit not yet refunded; disputed or held payments are handled by our team. Billing and usage records, which hold no names, emails or content, are kept for 6 years, as tax and accounting law requires. Details are in the Privacy Policy.

Check it on every response

x-st-region: us

The region that served the request. Every API response, errors included.

x-st-data-retention: none

Prompts and completions from this request are not stored. Every API response, errors included.

x-st-request-id: 01K5Z3T8R2V9Q4M6N1P7B0C8XD

The request's id, for usage records and support. Inference responses.

Give each agent the narrowest key that does the job, for no longer than it needs.

# A sub-agent key: inference only, $2 a day, gone in 12 hours
npx -y scaledthought@0.6 keys create --name sub-agent --child \
  --scopes inference --budget-usd 2 --expires-in 12h

# Replace a key; the old one keeps working for 24 hours
npx -y scaledthought@0.6 keys rotate key_… --grace 24h

# This key may have leaked: revoke it
npx -y scaledthought@0.6 keys revoke-self

Reference: keys in the docs.

Every change is checked

Changes to our code and infrastructure go through pull requests that must pass the tests, secret scanning, dependency vulnerability checks and static analysis before they merge.

Nightly scans

Every night we scan our code, dependencies, container images and infrastructure configuration for known vulnerabilities. Findings get a fix-by date by severity: critical 7 days, high 30 days, medium 90 days, low 180 days.

Infrastructure as code

Our cloud infrastructure is defined in code. Before each deployment, policy checks refuse any resource outside the region, any publicly readable data store, and any database, bucket, cache, secret store or log bucket without the region's own encryption key.

Pinned dependencies

Container base images are pinned by digest and CI actions by commit. Dependency updates are proposed weekly and go through the same checks.

Secrets

Service credentials live in each region's secret manager, never in code. Our CLI never takes a secret as a flag.