Resources

The US federal court system runs on CM/ECF — a distributed infrastructure where each of the 94 district courts, 94 bankruptcy courts, and 13 courts of appeals operates its own independent instance. There is no single federal court database and no official programmatic access layer.

PACER provides human-facing access on top of CM/ECF; turning that into structured data means parsing HTML across hundreds of independently maintained systems and normalizing it into a consistent schema. The Second and Ninth Circuits have migrated to ACMS, a newer appellate case management system with a different data structure; DocketLayer handles both transparently.

State court data operates without a federal equivalent of PACER. Each state — and in many cases each county within a state — maintains its own portal. DocketLayer covers general jurisdiction courts across all 50 states through integrations against statewide unified portals, Tyler Technologies Odyssey deployments across Illinois, Indiana, Minnesota, Missouri, and others, authenticated session portals such as West Virginia's WVPASS, and independent county systems.

Coverage is full for most states, partial where portal access is limited by the source system, and gated pending commercial approval in Texas. The state court data article and the coverage status article describe what each coverage level means for your queries.

Canadian coverage spans 25 courts: the Federal Court of Canada, the Federal Court of Appeal, the Tax Court of Canada, and 22 provincial and territorial superior courts.

Ten of those courts carry full coverage; the remainder are partial, reflecting the limits of each province's public portal. Federal courts and courts in officially bilingual provinces return data in English or French via the language parameter. Quebec's superior courts are integrated and indexed but gated pending a commercial agreement with SOQUIJ, Quebec's provincial data broker.

DocketLayer's payment model is worth understanding before you build. The x402 protocol makes payment the authentication mechanism — there are no API keys to provision, no accounts to create, and no subscriptions to manage.

Every request is paid at the point of query: $0.99 in USDC on Solana, settled through a four-step HTTP handshake.

Solana's sub-cent transaction fees make $0.99 micropayments economically viable; USDC keeps the query price stable regardless of market conditions. The payment articles cover the x402 handshake in detail, how USDC works as an SPL token on Solana, and how per-request billing compares to subscription pricing.

DocketLayer is built to operate without human intervention, and the articles in this section address what that requires in practice.

The MCP endpoint at api.docketlayer.ai/v2/mcp exposes seven tools — three free, four paid via x402 — that any MCP-compliant AI client can discover and invoke directly.

The article on designing APIs for agents covers the specific ways agent-native APIs differ from developer-facing ones: schema predictability, machine-readable error codes, idempotency, capability discovery, and payment as protocol. Whether you are integrating via MCP or the REST API, these articles describe the design decisions behind DocketLayer and what they mean for autonomous workflows.

These articles address specific professional workflows where court docket monitoring solves a concrete operational problem.

Each assumes you have a case number — discovery happens elsewhere. DocketLayer is the monitoring layer: structured, continuous, and per-query.