The global filings API.
46,243 companies, 21.5 million filings, 105 markets — through one typed REST and GraphQL endpoint. Markdown by default. Webhooks fire under 90 seconds.
What the API gives you.
Four primitives. Every filing in the index, callable in production, in the format your stack actually wants.
Markdown by default.
A 412-page PDF becomes ~3,200 tokens of Markdown. Tables intact, footnotes attached, ~70 % token reduction.
REST and GraphQL.
Typed schema. Same index, two query styles. JSON, Markdown, or original PDF on every endpoint.
Push, don't poll.
A new filing fires a webhook in under 90 seconds. Pipelines run on the event, not on the cron.
Free for LLM clients.
An MCP server every LLM client can speak. Filings inside Claude, ChatGPT, or any compliant agent — for free.
How it works.
From regulator publication to production-ready record. Four steps, end-to-end, in seconds.
Watch the regulators.
Continuous polling of every regulator and exchange in our coverage. New filings detected within seconds of publication.
Pull, dedupe, store.
Fetch the original document, deduplicate against the index, store the canonical PDF in our object store with audit trail.
Normalise to Markdown.
Tables, footnotes, headings, and identifiers extracted into a single canonical record. JSON, Markdown, and original PDF served from one record.
Push to your stack.
REST + GraphQL pull. Webhook push. S3 mirror, Snowflake, or Databricks shares for bulk consumers.
Try it in two lines.
Pull the latest filing for any ticker. Same endpoint serves Markdown, JSON, or the original PDF.
Use it inside your LLM client.
A free, open-source MCP server. 60-second setup in any compliant client. Three popular ones below.
Claude Desktop
Add a single block to claude_desktop_config.json. Filings become callable in any conversation.
Claude Code
One claude mcp add command. The agent has the index in-tool while you build.
ChatGPT
Custom GPT or Connectors. Surface every filing inside ChatGPT for research, analysis, and drafting.
Setup guide →Who builds with it.
Four audience views with sector-specific examples, customer references, and starter templates.
Alpha generation.
Fundamental signals from filings, normalised to feed quant models without scraper maintenance.
Read more → For asset managersPortfolio screens.
Filing-driven screens that update on the wire. Discretionary teams stop running 24-hours-late.
Read more → For analytics platformsEmbedded data layer.
License the index for redistribution. White-label or co-brand the filings layer in your product.
Read more → For academic researchCross-market studies.
Pre-cleaned, identifier-resolved filings for longitudinal and cross-market studies. Sandbox available.
Read more →Frequently asked.
Answers to the questions sales gets every week. If yours isn't here, the form on the next page goes straight to a data engineer.
Markdown filings, in your stack.
Talk to a data engineer about volume, regions, and integration. We reply within 24 hours.