ARDCatalog Lab

AGENTIC RESOURCE DISCOVERY · DRAFT v0.9

Give agents a map
to what you built.

Generate and validate the ai-catalog.json manifest that lets ARD registries discover MCP servers, A2A agents, Skills, APIs, and workflows.

Local validationURN checksValue-or-referenceQuery quality
INPUT

ai-catalog.json

CONFORMANCE REPORT

Discovery readiness

—
Waiting for a catalog

Paste JSON or load the official-style sample.

MINIMAL MANIFEST

Generate an ARD entry

Describe one resource, then extend the generated entries array for the rest of your catalog.

WHAT GETS CHECKED

Discovery metadata needs more than valid JSON.

01

Identity

Validates urn:air: identifiers, publisher-domain anchoring, host identity, duplicate IDs, and stable names.

02

Artifact envelope

Requires a media type and exactly one of url or inline data, with HTTPS guidance for referenced artifacts.

03

Search quality

Reviews descriptions, capabilities, tags, and the 2–5 representative natural-language queries used for semantic ranking.

04

Publishing

Checks spec version, timestamps, trust metadata shape, secret-like values, and well-known hosting requirements.

PUBLISHING FLOW

From catalog to discovery.

  1. 1
    Create

    List each callable resource with a stable identifier and artifact reference.

  2. 2
    Host

    Serve JSON at /.well-known/ai-catalog.json over HTTPS.

  3. 3
    Allow crawling

    Return application/json and Access-Control-Allow-Origin: *.

  4. 4
    Verify

    Run official conformance tooling as the final source of truth while ARD evolves.

ai-catalog.json generatorCreate an ARD manifest →ai-catalog.json validatorFind schema and quality issues →ARD publish checklistPrepare the well-known endpoint →

PURPOSE & METHODOLOGY

A focused tool, with the context to use it well

ARD Catalog Lab was created to validate and generate ai-catalog.json documents for Agentic Resource Discovery. It is intended for site owners making resources discoverable to agents through structured metadata. The tool is free, requires no account, and is paired with original explanations so you can understand the result rather than copy an unexplained output.

How to use the result

Start with a sample or a non-sensitive copy of your data, run the check, and review every finding before changing a production project. The validator combines syntax checks with actionable publishing guidance. Examples and checklists show how catalog fields connect to public resources and crawlable site architecture. Specifications and software evolve, so confirm high-impact decisions against the linked primary documentation.

Privacy and limitations

AI catalog metadata is processed in your browser wherever the tool page states that local processing is used. Avoid entering passwords, access tokens, personal data, or confidential material. Automated output may be incomplete because it cannot know your entire deployment, threat model, or organizational policy.

Independent and transparent

This is an independent developer resource, not an official certification service. Source code and issue tracking are available on GitHub. Read more about the project, review the privacy policy, or report an error.