KeyVex

get_nonprofit_filings

Read-only

Returns IRS Form 990-series e-filing records — the registry of every e-filed nonprofit return the IRS has released, 2017→present (~5.5M filings; ~400-750K/yr). One record per return: EIN, organization name, return type (990 = full; 990EZ = small; 990PF = private foundation; 990T = unrelated business income; the 2019-era index also carries IRS codes 990EO/990O verbatim), the tax period covered (YYYYMM), and the IRS release year. Use this when the user asks: does nonprofit X file with the IRS, when did a foundation last file, which returns has an EIN filed, or to anchor a nonprofit's identity (EIN) before joining grants / lobbying / OIG data by name. Records carry the filing's extracted FINANCIALS: total_revenue, total_expenses, total_assets_eoy, net_assets_eoy, and officers[] — top 25 by reported compensation with name/title. Coverage (reconciled 2026-07-08): 100% for release years 2019-2026, 97% for 2018, 69% for 2017 — the shortfall is IRS-side (the pre-2017 XML archives that held those filings' documents were retired by the IRS; the registry rows remain, without financials). Meanings follow the form: for 990T, total_revenue is unrelated-business taxable income. Officer compensation is as reported to the IRS. sub_date is set only for 2019-era records (later IRS indexes carry only the year — KeyVex never fabricates dates); tax_period is the reliable time axis. A nonprofit's fiscal year varies — tax_period 202506 means the period ENDING June 2025. Pure-publisher posture: the IRS index rows as published.

Input Schema

TableJSON Schema

Name	Required	Description
`ein`	No	Employer Identification Number (any format — digits extracted).
`name`	No	Case-insensitive substring against the organization name.
`limit`	No	Maximum records. Default 50, max 500.
`since`	No	tax_period lower bound — 'YYYY-MM' or 'YYYYMM' (inclusive).
`until`	No	tax_period upper bound — 'YYYY-MM' or 'YYYYMM' (inclusive).
`object_id`	No	Direct lookup by IRS OBJECT_ID.
`sort_order`	No	Sort by tax_period. Default desc.
`return_type`	No	IRS code verbatim: '990', '990EZ', '990PF', '990T' (also '990EO'/'990O' in 2019-era records).
`submission_year`	No	IRS release/index year (2017→present; financials partial for 2017 — IRS retired the pre-2017 XML archives).

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 4, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover the read-only/safe profile, yet the description adds substantial context: coverage percentages reconciled by date, why 2017 financials are partial (IRS retired pre-2017 XML), sub_date being set only for 2019-era records, and a 'never fabricates dates' guarantee. This is exactly the behavioral detail annotations cannot convey.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with what it returns before the usage and caveat material. It is long (~350 words) with several parenthetical asides, but nearly every sentence carries distinct information needed by a data-catalog consumer with no output schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema and 9 parameters, the description fully compensates: it enumerates returned fields (EIN, name, return_type, tax_period, release year, financials, officers), explains semantics per form type, and documents coverage limits. An agent has everything needed to call and interpret results.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3. The description goes beyond it by decoding return_type values (990 full, 990EZ small, 990PF foundation, 990T UBI), explaining tax_period as YYYYMM with a fiscal-year ENDING example, and clarifying that since/until bound tax_period.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Returns IRS Form 990-series e-filing records') and immediately scopes it (2017→present, one record per return). It also positions the tool against siblings by describing its role as an EIN identity anchor 'before joining grants / lobbying / OIG data by name,' so an agent can tell it apart from those tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit, concrete when-to-use scenarios ('does nonprofit X file with the IRS, when did a foundation last file, which returns has an EIN filed'). It does not name an alternative tool or state when-not-to-use, but the use-case framing is strong and unambiguous.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All KeyVex tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?