Acceptable use · 2026-08-18
Use content, APIs, feeds and artifacts only within the permissions, volume, purpose, territory and term shown in the record or signed agreement. Retrieval permission is not model-training, redistribution or commercial-republication permission.
Preserve attribution, canonical links, disclosures, content type, source context, version identifiers and rights metadata. Do not present sponsored material as independent reporting or opinion as straight news.
Consume correction and change feeds, refresh stale copies, and stop using withdrawn or expired material when the record requires it. You may not suppress an update because an older version is more commercially useful.
Do not use the service for unlawful surveillance, targeted harassment, discrimination, doxxing, source identification, impersonation or material deception. Do not combine records to defeat privacy protections or create high-impact decisions about a person without appropriate review.
Do not share credentials beyond the authorized project team, ship keys in client code, probe for vulnerabilities without permission, bypass quotas, create excessive accounts, interfere with delivery, scrape restricted fields or automate malicious traffic.
Evaluation data may support internal retrieval, citation and embedding tests. It may not train or fine-tune a model. Agents must distinguish retrieved reporting from generated analysis and link users to the canonical article when presenting National Tribune material.
We may throttle, suspend or revoke access; require deletion or correction; preserve evidence; and pursue contractual or legal remedies. Urgent security or rights risks may be acted on without advance notice.
