Give AI a schema. Get predictable JSON.

Define the fields and data types once. Cuddler carries that exact shape from the published contract to the AI response and validates it before your application uses it.

1 Define Set the expected shape
The Artifact Specification and one Document Role turn your requirements into explicit fields, types, and rules.
{ fields, types, rules }
2 Publish Create the contract
A definition-building agent publishes an Artifact Definition with the exact Data Schema and guidance.
Artifact Definition + Data Schema
3 Generate Get predictable JSON
A document-building agent follows the schema and returns the fields and data types your code expects.
{ "expected": "shape" }
4 Validate Use only valid data
Validation blocks mismatched responses. Your application consumes the JSON only when the contract passes.
✓ Schema valid

The schema is the handoff. AI follows it, validation checks it, and your application can rely on it.

Build an Artifact Definition Open the AI implementation guide

What becomes easier to control

Make hidden constraints visible

Encode required fields, types, formats, enumerations, cardinalities, and conditional rules so engineers can inspect the intent before automation scales it.

Keep instructions with the contract

Carry schema-node for-ai guidance with the published definition so compatible assistants and human reviewers work from the same instructions.

Preserve meaning across tools

Keep governing standards, exact schema identities, publication inventories, examples, and implementation files aligned so a handoff does not silently reinterpret the document.

Know which layer failed

Validate Data and Template independently so a mismatch stops at the responsible layer before it becomes a broken rendered document.

Implementation stories from real document workflows.

Each story connects an engineering pressure to the contract, validation boundary, and review path used to address it. The source note on each page states the evidence boundary.

Boostware used Cuddler to build a workflow system around document generation, turning operational know-how into structured, reviewable outputs that fit human approval gates.

Read the story

Cocoon CS used Cuddler to define reusable document templates for evidence collection, helping audit-facing outputs stay consistent across frameworks, controls, and review cycles.

Read the story

Day Guardian used Cuddler to build child care safety reports that stayed structured, traceable, and ready for review inside a broader compliance reporting system.

Read the story

EZ Support used Cuddler to streamline the development of SOPs autonomously, giving AI a controlled publishing contract for the procedures that support secure IT operations.

Read the story

What do you want to do?

Create a new contract, use one that already exists, or connect Cuddler to your agent or integration. Pick the path that matches what you're working on.

Create a contract Give your agent the Artifact Specification, the right Document Role Specification, your source material, and a few examples. It can use them to build a definition bundle you can inspect. Create an Artifact Definition

Use an existing contract Choose an Artifact Definition from the library and use its exact schema and built-in guidance to create documents you can validate. Browse the Library

Connect your tools Find the canonical URLs, current releases, downloads, and validation steps your agent or integration needs. View Integration Details

Cuddler publishes document contracts that give AI an exact Data Schema and give applications predictable, validatable JSON.

TrackThat Inc.

TrackThat Inc. is the corporate owner and accountable publisher of the Cuddler technology specification.

Visit TrackThat

Explore

Standards

Learn

License

Creative Commons logomark

You are free to share, use, adapt, and modify this material, including for commercial purposes, provided you give appropriate attribution and include a link to the original source URL and the CC BY 4.0 license .

License Scope

This license applies to the public Cuddler standards pages, downloadable JSON publications, public schema artifacts, public examples, and related downloads published on Cuddler.dev unless a page or file explicitly states otherwise.

Attribution Requirements

When you reuse or adapt public material from this site:

Identify TrackThat as the source.
Link to the original Cuddler.dev source URL you used.
Link to the official CC BY 4.0 license.
Indicate whether you made changes.

What Is Not Included

Cuddler, TrackThat, or other referenced marks, logos, and branding are not included in the CC BY 4.0 license unless a page explicitly says otherwise.

Terms of Service

Welcome to Cuddler.dev. These terms describe the rules for using the Cuddler.dev website, including its documentation, schema artifacts, examples, embedded guidance, and related downloads.

Updated: March 21, 2026

Cuddler.dev is owned and operated by TrackThat Inc. If you enter into a commercial subscription, services engagement, or other contractual arrangement connected to Cuddler.dev, your contracting counterparty is TrackThat Inc.

1. Agreement Acceptance

By accessing or using Cuddler.dev, you agree to these terms on your own behalf and on behalf of any person or organization you represent. If you do not agree, do not use the site or any materials published through it.

2. Permitted Website Use

2.1 General Use

You may use this website only for lawful purposes and only if you can form a legally binding agreement under the laws that apply to you.

2.2 Linking to Cuddler.dev

You may link to public Cuddler.dev pages in a fair and non-misleading way. You may not frame the website, imply endorsement that does not exist, or present Cuddler.dev content as your own.

2.3 Copying or Distributing Content

Public standards, schema artifacts, examples, and related downloads that link to the Cuddler license page are licensed under CC BY 4.0. Branding, marks, and other materials that do not carry that notice remain restricted unless TrackThat Inc. gives permission separately.

2.4 Accounts and Restricted Services

If any portion of the service requires credentials or controlled access, you are responsible for maintaining the confidentiality of those credentials and for all activity performed using them.

3. Personal Information

3.1 Information Accuracy

Any information you submit to Cuddler.dev must be accurate, current, and complete.

3.2 Privacy Practices

TrackThat Inc. collects and uses personal information in accordance with the Cuddler.dev Privacy Policy. By using the site, you acknowledge those privacy practices.

4. Feedback and Submissions

If you send feedback, suggestions, or non-confidential materials to Cuddler.dev, you grant TrackThat Inc. a non-exclusive, worldwide, royalty-free right to review, use, and incorporate that feedback to improve the site, service, and related materials.

5. Changes to Terms

TrackThat Inc. may revise these terms from time to time. Continued use of Cuddler.dev after a posted revision constitutes acceptance of the updated terms.

6. Accuracy and Availability

We aim to keep Cuddler.dev accurate and available, but documentation, examples, schema files, embedded guidance, and other site materials may contain errors, omissions, or outdated information. We may correct, revise, remove, or replace material at any time without notice.

7. Intellectual Property

Cuddler, Cuddler.dev, and associated logos and branding are owned by or licensed to TrackThat Inc. Public standards, schema artifacts, examples, and related downloads that link to the Cuddler license page are licensed under CC BY 4.0, but that license does not extend to Cuddler trademarks, logos, or branding unless a page expressly says otherwise.

8. Disclaimer and Limitation of Liability

Cuddler.dev is provided on an “as is” and “as available” basis. To the fullest extent permitted by law, TrackThat Inc. disclaims warranties and will not be liable for indirect, incidental, special, consequential, or punitive damages arising from or related to your use of the site or reliance on its materials.

Questions about these terms may be sent to hello@cuddler.dev.

Privacy Policy

Updated: March 21, 2026

Cuddler.dev is operated by TrackThat Inc. This policy explains how TrackThat Inc. collects, uses, stores, and protects information when you visit Cuddler.dev, request information, download public materials, or otherwise interact with the site.

1. Information We Collect

Contact information you choose to provide, such as your name, email address, company name, or message contents.
Technical and usage information such as IP address, browser type, device information, referring pages, on-site activity, and approximate session timing.
Operational records associated with requests for documentation, schemas, examples, embedded guidance, or support interactions.

2. How We Use Information

To operate, secure, and improve Cuddler.dev and its published materials.
To respond to inquiries, support requests, and commercial discussions.
To measure site usage, investigate abuse, and maintain service reliability.
To satisfy legal, contractual, security, and record-keeping obligations.

3. Cookies and Similar Technologies

Cuddler.dev may use cookies, logs, or similar technologies to support core site operation, performance analysis, and security monitoring. You can manage cookies through your browser settings, though some site features may not function as intended if those controls are disabled.

4. Sharing of Information

TrackThat Inc. does not sell your personal information. We may share information with service providers, hosting or analytics vendors, legal advisers, or other parties when reasonably necessary to operate Cuddler.dev, protect the service, comply with law, or support a transaction or engagement handled by TrackThat Inc.

5. Retention and Security

We retain information only for as long as reasonably necessary for business, legal, security, and operational purposes. TrackThat Inc. uses administrative, technical, and organizational safeguards appropriate to the nature of the information and the services provided, but no method of transmission or storage is completely secure.

6. International Processing

Your information may be processed in jurisdictions where TrackThat Inc. or its service providers operate. By using Cuddler.dev, you understand that your information may be transferred to and processed in those locations.

7. Your Choices

You may request access to, correction of, or deletion of personal information, subject to applicable legal and contractual limits.
You may opt out of non-essential communications by following the unsubscribe instructions included with those messages or by contacting us directly.
You may contact us to ask questions about how your information is handled by TrackThat Inc.

8. Policy Updates

TrackThat Inc. may update this Privacy Policy from time to time. Any revised version will be posted on Cuddler.dev with the updated effective date.

Privacy questions may be sent to hello@cuddler.dev.

Compliance

Cuddler.dev is operated by TrackThat Inc. This page summarizes the governance, privacy, security, and operational controls that support the site and any related commercial engagements handled by TrackThat Inc.

Updated: March 21, 2026

1. Governance

TrackThat Inc. maintains contractual, administrative, and change-management controls for Cuddler.dev and related deliverables. Commercial commitments, customer communications, and service obligations associated with Cuddler.dev are administered through TrackThat Inc.

2. Security Practices

Access to administrative systems and publication workflows is limited to authorized personnel.
Changes to public artifacts, documentation, and site behavior are managed through controlled source and deployment workflows.
Reasonable monitoring, logging, and incident-response practices are used to detect and investigate operational or security issues.
Sensitive data is handled using least-privilege and need-to-know principles where applicable.

3. Privacy and Data Handling

Personal information is processed only for legitimate business, operational, legal, or support purposes.
Retention is limited to the period reasonably necessary for those purposes.
Third-party services are evaluated for operational fit and appropriate handling of information shared with them.

4. Documentation Integrity

Cuddler.dev publishes the Specification Root, versioned Document Role, the shared Artifact Specification, Artifact Definition artifacts, and examples. Guidance for assistants is embedded inside the published specification and schema files. TrackThat Inc. uses source-controlled publication workflows to preserve traceability between maintained source content and public site output.

5. Customer and Vendor Review Support

Where appropriate for an active engagement, TrackThat Inc. may respond to reasonable diligence questions concerning Cuddler.dev operations, privacy practices, or service controls, subject to confidentiality, scope, and legal constraints.

6. Questions

Compliance, security, or diligence questions may be sent to hello@cuddler.dev.

AI Safety Policy

Cuddler.dev is operated by TrackThat Inc. This policy explains how AI-assisted systems, generated outputs, and automation patterns associated with Cuddler.dev are expected to be used, reviewed, and constrained.

Updated: March 23, 2026

1. Purpose and Scope

Cuddler.dev publishes standards, examples, schema artifacts, embedded guidance, and implementation guidance intended to support safe and structured AI-assisted document generation. This policy applies to AI-generated or AI-assisted content, workflows, and integrations that TrackThat Inc. publishes through Cuddler.dev or uses to operate related services.

2. Human Accountability

AI systems may assist with drafting, transformation, classification, or rendering workflows, but people remain responsible for decisions, approvals, and public publication. TrackThat Inc. expects human review before relying on AI-generated outputs for legal, commercial, compliance, or operationally sensitive use cases.

3. Validation Before Use

Generated JSON and templates should be checked against the applicable schema, specification, or implementation contract before use.
Validation failures, version mismatches, or unsafe render conditions should block downstream publishing or rendering.
Examples and reusable artifacts should prefer deterministic, inspectable structures over opaque free-form output.

4. Safety Controls for Rendered Output

Untrusted content should remain subject to strict sanitization by default.
Scripts, unsafe embeds, event handlers, and non-whitelisted protocols should not be introduced through AI-generated templates or content.
Any relaxation of sanitization or external-asset controls should be deliberate, reviewable, and appropriate to the trust level of the content source.

5. Data Handling and Confidentiality

Artifact Documents are confidential by default unless explicitly designated otherwise. Users should not place confidential, regulated, or sensitive information into AI systems unless the applicable service, contract, and safeguards are appropriate for that information. Data shared with AI-assisted workflows should be limited to what is reasonably necessary for the task.

6. Prohibited or Restricted Uses

Do not use Cuddler.dev materials to misrepresent authorship, provenance, approvals, or compliance status.
Do not use AI-generated output from Cuddler.dev workflows as a substitute for legal, security, financial, or other professional advice without qualified review.
Do not use published schemas, examples, or embedded guidance to facilitate harmful, deceptive, abusive, or unlawful activity.

7. Monitoring, Correction, and Improvement

TrackThat Inc. may review feedback, investigate unsafe behavior, revise published materials, or remove content when accuracy, safety, or misuse concerns arise. Safety-related fixes may include prompt changes, schema changes, documentation changes, stricter validation, or tighter publication controls.

8. Questions and Incident Reporting

Questions about this policy, or reports of unsafe AI behavior associated with Cuddler.dev, may be sent to hello@cuddler.dev.

Stop making AI invent document rules

Give AI a schema. Get predictable JSON.

What becomes easier to control

Make hidden constraints visible

Keep instructions with the contract

Preserve meaning across tools

Know which layer failed

Inspect the contract behind every Artifact Definition

Implementation stories from real document workflows.

What do you want to do?

Explore

Standards

Learn