Mozilla Data Collective

Search datasets

search

Read-onlyIdempotent

Search the Mozilla Data Collective catalog of AI training datasets by natural-language query, optionally narrowed by task, language, license, format, price, sample availability or publish date. Returns matching datasets as {id, title, url}; pass an id to the fetch tool for full details. Call list_filters first if you intend to filter — filter values must match the catalog exactly.

Input Schema

TableJSON Schema

Name	Required	Description
`sort`	No	Result ordering. Defaults to 'relevance'; use 'newest' or 'size' only when the user asks for it.
`task`	No	Restrict to these machine-learning tasks, e.g. ['ASR', 'TTS'].
`limit`	No	Maximum number of results to return (1-25).
`query`	Yes	Natural-language search query describing the datasets you are looking for, e.g. 'Spanish speech recordings for TTS training'. Descriptive phrases retrieve better than single keywords.
`format`	No	Restrict to these file formats, e.g. ['WAV', 'MP3']. Values must match exactly (case-sensitive); call the list_filters tool to get the valid ones.
`isPaid`	No	true returns only paid datasets, false only free ones. Omit to include both.
`locale`	No	Restrict to these language/locale codes, e.g. ['sw', 'pt-BR']. Values must match exactly (case-sensitive); call the list_filters tool to get the valid ones.
`license`	No	Restrict to these license abbreviations, e.g. ['CC0-1.0', 'CC-BY-4.0']. Values must match exactly (case-sensitive); call the list_filters tool to get the valid ones.
`hasSample`	No	true returns only datasets that publish a downloadable sample, useful when the user wants to try data before committing. false behaves the same as omitting it.
`uploadDate`	No	Restrict to datasets published within this recent window.
`sortDirection`	No	Direction for the sort field. Only meaningful alongside sort='newest' or sort='size'.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

First observedSep 7, 2026

TDQS

A4.5/5.0

Behavior4/5

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

Annotations already provide readOnlyHint, openWorldHint, idempotentHint, and destructiveHint. The description adds meaningful behavioral context beyond those: the exact return shape ({id, title, url}) and the exact-match requirement for filter values, which is important operational detail for the agent.

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

Conciseness5/5

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

Three sentences, each earning its place: what the tool does, what it returns and how to continue with fetch, and the critical precondition for filtering. The information is front-loaded and tightly written.

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?

Despite 11 parameters, the schema fully covers them, and the description covers the essential non-schema context: return format, follow-up tool routing, and exact filter matching. With no output schema present, the explicit return shape fills the gap. Nothing critical is missing for an agent to invoke this correctly.

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

Parameters3/5

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

Schema description coverage is 100%, so the schema already documents all 11 parameters thoroughly. The description's filter list adds a useful high-level summary but does not meaningfully enhance individual parameter semantics beyond what the schema already provides.

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?

The description clearly states a specific action ('Search'), a specific resource ('Mozilla Data Collective catalog of AI training datasets'), and the natural-language query mechanism. It also distinguishes itself from siblings by noting that matching dataset ids go to fetch and that list_filters should be called for filter values.

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

Usage Guidelines5/5

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

The description gives explicit when-to-use guidance: use natural-language search to find datasets, call list_filters before filtering, and pass an id to fetch for full details. This effectively routes the agent between search, list_filters, and fetch without ambiguity.

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

All Mozilla Data Collective tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.5/5.0

Disambiguation5/5

Each tool has a clearly distinct purpose: list_filters enumerates accepted filter values, search returns matching dataset summaries, and fetch retrieves full details for a single dataset. There is no meaningful overlap or ambiguity between them.

Naming Consistency4/5

All names are lowercase imperative verbs and are easy to understand. list_filters follows a verb_noun pattern while search and fetch are single-word verbs, but the inconsistency is minor and does not hurt usability.

Tool Count5/5

Three tools is an appropriate, well-scoped size for a read-only catalog search and retrieval server. Each tool serves a necessary part of the discovery workflow: filter discovery, search, and detail fetch.

Completeness5/5

The tool surface fully covers the stated domain: discover available filters, search the catalog, and fetch detailed dataset information. There are no obvious dead ends or critical missing operations for this read-only purpose.

Resources

Need Help?