> ## Documentation Index
> Fetch the complete documentation index at: https://vaquill.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Roadmap
> Corpora Vaquill AI is building next, what is deliberately out of scope, and how to detect new coverage programmatically
What is coming to the corpus, and what is not.
Everything here ships through the same pipeline as the current data: official government source, chunked and embedded, citation-linked, on a recurring refresh. Each lands behind the same `corpusType` filter, so there is nothing to integrate.
Queryable today: [Coverage](/docs/api-guide/coverage). Already shipped: [Changelog](/docs/api-guide/changelog).
Dates are targets, not commitments. Poll `GET /statutes/coverage` to detect new corpora rather than tracking this page, and see [Coverage monitoring](/docs/api-guide/recipes/coverage-monitoring) for a scheduled diff.
## Completing existing coverage
Three corpus types are live but not yet across all 52 jurisdictions.
All are rolling out state by state on the same pipeline and the same refresh cadence as the states already covered.
| Corpus | Today | Target |
| ----------------------------------------------- | ---------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| State regulations (`REGULATION`) | 46 states | All 52 jurisdictions |
| State court rules (`STATE_RULES`) | 42 jurisdictions | Broad coverage. The remaining states either publish no consolidated rules text from a government source, codify their rules into the statute book (already served under `STATE`), or restrict automated access by published court policy. |
| State agency guidance (`STATE_AGENCY_GUIDANCE`) | 49 jurisdictions, 12,808 documents | Effectively complete; three jurisdictions publish no browsable bulletin archive |
State agency guidance is each insurance regulator's own bulletins and circular
letters, the state-level analog of federal `AGENCY_GUIDANCE`. Three
jurisdictions publish no cleanly scopable insurance-only archive and are not
currently planned: Pennsylvania, Colorado and Louisiana.
Pennsylvania's Consolidated Statutes are complete; its older unconsolidated (Purdon's) statutes are a separate later addition, prioritized for backfill.
## Federal law and court rules
Targeted to land by the **start of Q4 2026**, with the exception noted below.
Federal local rules are the largest item in this table by a wide margin,
since they mean roughly 200 separate courts each publishing their own set,
mostly as PDFs, with no common schema. Expect it to trail the other three.
| Source | What it adds | Official source |
| --------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------- |
| Federal Register notices | The third Federal Register document class. We carry final and proposed rules today; notices cover meeting announcements, information collections, agency determinations, and grant availability. | govinfo, federalregister.gov |
| Public Laws / Statutes at Large | Slip laws as enacted, before codification into the US Code. Public law numbers appear as a metadata field today, but the acts themselves are not served as documents. | govinfo |
| OLC and Attorney General opinions | Office of Legal Counsel (OLC) opinions plus formal Attorney General (AG) opinions, the executive branch's own binding legal interpretations. | justice.gov/olc |
| GAO decisions | Government Accountability Office (GAO) decisions of the Comptroller General on appropriations law and bid protests. | gao.gov |
| Federal local rules | The local rules of the 94 district courts, the 13 circuits, and the bankruptcy courts, adopted under 28 U.S.C. 2071 and Federal Rule of Civil Procedure 83. These modify the national Federal Rules in every district, so the national rules alone are an incomplete picture of what governs a filing. Citable in their own form, for example `N.D. Cal. Civ. L.R. 7-2`. | uscourts.gov and each court's own site |
## Tax
Two additional Internal Revenue Service (IRS) corpora, beyond the IRS notices, revenue procedures, revenue rulings and announcements already served under agency guidance.
| Source | What it adds |
| ---------------------- | ---------------------------------------------------------------------------------- |
| Chief Counsel Advice | Written determinations issued by IRS Chief Counsel to field offices. |
| Private Letter Rulings | Taxpayer-specific rulings on the application of tax law to a proposed transaction. |
## Administrative adjudication
A new corpus type: the record of what federal agencies actually decided, as opposed to what the rules say.
These are public domain under 17 U.S.C. 105 and are genuinely under-aggregated, since each body publishes into its own silo with no shared format or API.
| Body | What it adds |
| ------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| SEC (Securities and Exchange Commission) administrative proceedings | The full order stream, including Accounting and Auditing Enforcement Releases (AAERs) and Rule 102(e) accountant disqualifications. Roughly 18,400 proceedings since the early 2000s. |
| DOL (Department of Labor) OALJ, ARB and BRB | Decisions of the Office of Administrative Law Judges (OALJ), the Administrative Review Board (ARB) and the Benefits Review Board (BRB), across an 80+ case-type taxonomy (SOX, AIR21, STAA, ERA, OSH, OFCCP, LCA, black lung, longshore). |
| NLRB (National Labor Relations Board) | Administrative law judge and Board decisions. Extends the NLRB General Counsel and Division of Advice memoranda already in the corpus. |
| MSPB (Merit Systems Protection Board) | Precedential and nonprecedential decisions. |
| FDIC, OCC and Federal Reserve | Enforcement actions and consent orders from the Federal Deposit Insurance Corporation, the Office of the Comptroller of the Currency and the Federal Reserve. Extends the guidance from these agencies already in the corpus. |
On customer demand: the Occupational Safety and Health Review Commission (OSHRC), the Federal Mine Safety and Health Review Commission (FMSHRC), the Federal Energy Regulatory Commission (FERC), Commodity Futures Trading Commission (CFTC) reparations, the Board of Immigration Appeals (BIA), the US Tax Court, the Armed Services and Civilian Boards of Contract Appeals (ASBCA and CBCA), the EPA Environmental Appeals Board, Federal Trade Commission (FTC) administrative decisions, and the Departmental Appeals Board (DAB) of the Department of Health and Human Services (HHS).
## Federal regulatory dockets
Rulemaking dockets and the public comment record behind a rule, sourced from three official, documented government APIs.
| Source | What it adds |
| ------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| regulations.gov | Federal rulemaking dockets, documents, and public comments across participating agencies. |
| FCC ECFS | Telecom proceeding filings from the Federal Communications Commission (FCC) Electronic Comment Filing System (ECFS), with filers, documents and submission types. |
| SEC EDGAR comment letters | Securities and Exchange Commission staff comment letters (`UPLOAD`) and filer correspondence (`CORRESP`), from the Electronic Data Gathering, Analysis and Retrieval system (EDGAR). |
## Agencies not yet covered
No documents from these agencies are in the corpus today.
Counts are the document categories planned for each.
| Agency | Categories | Domain |
| ------------------------------------------------ | ---------- | --------------------------------------------------- |
| Federal Trade Commission (FTC) | 4 | Consumer protection and competition |
| Department of Justice (DOJ) Antitrust Division | 1 | Merger and conduct enforcement |
| Securities and Exchange Commission (SEC) | 3 | Securities regulation and disclosure |
| Customs and Border Protection (CBP) | 3 | Import classification, valuation, and admissibility |
| Consumer Financial Protection Bureau (CFPB) | 2 | Consumer financial protection |
| Department of Labor (DOL) Wage and Hour Division | 2 | Wage, hour, and leave enforcement |
## Other federal guidance
| Source | Domain |
| -------------------------------------------------------------------------- | ------------------------------------------------------ |
| Centers for Medicare & Medicaid Services (CMS) Medicare Manuals | Medicare coverage, billing, and program administration |
| Department of Justice (DOJ) Americans with Disabilities Act (ADA) guidance | Disability access and accommodation |
| Equal Employment Opportunity Commission (EEOC) guidance | Employment discrimination |
| Food and Drug Administration (FDA) guidance documents | Drug, device, and food regulation |
| FDA warning letters | Enforcement correspondence to regulated entities |
| Department of Housing and Urban Development (HUD) fair housing guidance | Housing discrimination |
| Office of Foreign Assets Control (OFAC) FAQs | Sanctions compliance |
## Out of scope
Deliberate exclusions, not gaps. If you need one of these, we are not your supplier.
| Not building | Why |
| ---------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| General case law and court dockets | We are a primary-law corpus: statutes, regulations, constitutions, court rules, agency material. |
| Judge-specific procedural material (standing orders, chambers procedures, individual practice rules) | See the court-material line below. |
| Practice notes and practice directions | A Commonwealth instrument with no direct US federal equivalent. The closest analogues are local rules and court-wide general orders. |
| Municipal and county ordinances | Public-domain text, but locked in three commercial publisher platforms with no comprehensive source. |
| Standards incorporated by reference into the CFR (ASTM, NFPA, ISO) | Published by private standards bodies. Courts have held only non-commercial dissemination is fair use, which is incompatible with a paid API. |
| NAIC model law text | A copyrighted, commercially sold publication, not open government primary law. Each state's own insurance bulletins are in scope and are now covered across 40 jurisdictions under `STATE_AGENCY_GUIDANCE`. |
### The line on court material
**We carry rules a court formally adopts and publishes as binding on everyone appearing before it. We do not carry an individual judge's preferences.**
Formally adopted rules are stable, citable, published under statutory authority, and amended on a visible cycle. That covers the Federal Rules and Supreme Court Rules today, the state court rules in the [Coverage](/docs/api-guide/coverage) matrix, and the federal local rules above.
Judge-specific material fails every one of those tests: over 1,500 judges, no consistent location, no citation form, revised or voided silently on rotation. We could ingest it. We could not keep it honest, and a confidently stale answer is worse than none.
## Related
Everything queryable today, with live per-jurisdiction counts.
What has already shipped, in date order.
Poll `/statutes/coverage` on a schedule and diff it, so you detect new corpora automatically.
What each `corpusType` token holds.