API documentation

Sanctions screening API usage and allowance limits

Read your used, reserved, and remaining screening allowance. Count completed subjects across API requests, batches, dashboard checks, and monitoring.

Updated

250 free live screenings. One lifetime allowance. No credit card required.
Unit
Completed subject

Includes individual checks, batch rows, and rescreens.

Read usage
GET /usage

Inspect used, reserved, allowance, and remaining.

At the cap
Screening stops

No automatic overage charges are created.

Available capacity = effective limit − used − reserved

Used
Completed individual checks, batch rows and monitoring rescreens. Idempotent retries do not count twice.
Reserved
Accepted work that has not yet completed. Cancellation and failures release unused reservations.
Remaining
The nonnegative capacity available now. Monitoring forecasts estimate future work; they do not reserve it.

The unit is a completed subject

Each successfully completed production screen counts one subject, whether the outcome is potential_match or no_match. Each completed batch row counts separately. Each completed monitoring rescreen counts again because it is a new check. HTTP requests, source count, and number of potential matches do not directly define the billable unit.

Idempotent retries do not count twice. Failed rows and failed screening attempts do not count as completed screens. A fresh request with a new idempotency key is a new screen, even if the same subject was screened before. A cached idempotent response is the original operation, not new usage.

Read allowance before accepting more work

Use usage:read. GET /usage returns data.environment, period, used, reserved, limit, planLimit, spendingCap, effectiveLimit, remaining, plan, events, fileStorage, monitoringForecast and forecastUnavailable. These values are a point-in-time view; a concurrent job can consume remaining capacity before your next request.

remaining is max(0, effectiveLimit - used - reserved). effectiveLimit respects a positive configured subject cap as well as the plan allowance. limit and planLimit describe the plan capacity; do not treat either as remaining work. monitoringForecast can be null with a forecastUnavailable reason. File storage values use bytes and are separate from subject usage.

A batch reserves capacity before processing, so used alone can understate committed work. Failed rows and cancellation release unused reservations; completed rows and successful rescreens count once. On usage_cap_reached, inspect reserved work and the period/cap rather than retrying immediately. On rate_limited, slow HTTP requests; that error does not mean the monthly subject allowance is exhausted.

Terminal · curl
curl --fail-with-body "https://www.sanctionskit.com/api/v1/usage" \
  -H "Authorization: Bearer $SANCTIONSKIT_API_KEY"

Billing periods and organization caps

Your first workspace gets 250 free live screenings for its lifetime, shared by the API and dashboard. This allowance has no monthly reset and needs no card. Paid plans use monthly billing periods. Optional sandbox tests have a separate allowance that resets each UTC calendar month. Sandbox checks are never billed.

The settings field spendingCap means a maximum subject count, not a dollar-denominated overage budget; zero uses the plan allowance. There are no automatic overages. Read the current effective limit and remaining capacity using the request above.

Budget monitoring explicitly

One daily monitor can use roughly 30 subjects in a 30-day period before source-update rescreens. The monitoring and usage forecast estimates scheduled volume through the current period, using the lower of plan allowance and any configured subject cap, minus used and reserved subjects. It shows projected exhaustion and a cadence preview but does not reserve future capacity. Source changes, retries, future batches and manual screens are excluded. If the cap is reached, screening is blocked without creating extra charges.

File storage is a separate allowance

The free lifetime offer includes 100 MiB of private file storage. Starter includes 1 GiB, Growth 10 GiB, and Scale 50 GiB per organization. Sandbox has a 100 MiB sublimit within the workspace allowance. Private bulk screening uploads, supporting text attachments, and in-progress upload reservations use this capacity. This is storage capacity, not a monthly upload allowance. Database screening evidence and external document links do not count toward it.

Each batch upload is limited to 25 MiB and 10,000 rows. Supporting case documents accept UTF-8 .txt or .md files up to 200,000 bytes. When the allowance is full, new files are rejected without deleting existing evidence or adding charges. Upgrading to a plan with more storage increases capacity; separate self-service storage add-ons are not available.

Original PDFs, images, and passport scans are linked, not uploaded to or stored by SanctionsKit. Upload them directly to your organization’s document service and add the existing HTTPS link to the case. Access, security, and retention remain governed by that service and your configuration. Owners and administrators manage approved services in Settings → Organization documents; see organization controls.

Request and response reference

Base URL: https://www.sanctionskit.com/api/v1. Open an operation for its complete headers, body fields, response formats, examples and error codes.

Browse the complete API reference or download the OpenAPI contract.

Frequently asked questions

What counts as one billable screening subject?

One successfully completed production screening counts once, whether it returns a potential match or no-match. Each completed batch row and monitoring rescreen counts separately. HTTP request count, number of sources, and number of returned candidates do not define the unit.

Why should I include reserved usage when planning a batch?

A batch reserves capacity before it finishes. used alone can understate the work already accepted. Read reserved and remaining too. Failed or cancelled rows release unused reservations, while completed rows still count toward the shared allowance.

Is spendingCap an automatic overage budget in dollars?

No. It is a maximum subject count. A positive cap can reduce the effective plan allowance; zero uses the plan allowance. There are no automatic overages. Read effectiveLimit and remaining to understand how much work can be accepted.

Does the monitoring forecast reserve future screening capacity?

No. It estimates scheduled volume in the current period using the available allowance. It excludes source-change activity, retries, future batches, and manual screens. Review the forecast alongside your other planned work; a later request can still hit the cap.

KEEP BUILDING

Where to go next