Singularity Layer MCP

create_dataset_with_payment

Step 2 of buying a training dataset: submit the job again with the signed X-Payment payload and the quote id. Returns in seconds with a dataset id and a claim token; generation then runs in the background. SAVE THE CLAIM TOKEN — it is shown once and is the only way to read the job or download the result.

Input Schema

TableJSON Schema

Name	Required	Description
`name`	No	Name for the dataset
`rules`	No	House rules the assistant must NEVER break, e.g. "Never promise a refund". Without them the model invents its own policy, differently each time
`verify`	No	Check every row with a second model and regenerate any that break the rules. ON by default, priced separately ($0.10 per 100 managed, $0.03 per 100 grid where the grid checks its own work so nothing leaves the network). false = skip checking, pay generation only
`modelId`	No	Exact managed model from list_dataset_pricing (decides the tier and price)
`network`	No	Chain you will pay on (default base)
`quoteId`	Yes	quote_id from request_dataset_payment
`gridModel`	No	Model on the decentralized encrypted grid: cheaper, slower, data never leaves the network
`modelTier`	No	Managed model speed/quality tier. Omit when using gridModel
`targetRows`	Yes	How many examples to generate (50-2000). Price scales with this
`webhookUrl`	No	Public HTTPS URL to notify when the dataset is ready; the POST is signed
`description`	Yes	What the model being trained should do, in one plain sentence
`seedExamples`	Yes	5 to 20 example conversations showing the behaviour you want
`systemPrompt`	No	Standing instruction every generated example should follow
`paymentPayload`	Yes	Raw X-Payment payload from the paying client, as JSON or a base64 string

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedSep 3, 2026

addedInput schema / properties / verify

Added value: +{
+  "description": "Check every row with a second model and regenerate any that break the rules. ON by default, priced separately ($0.10 per 100 managed, $0.03 per 100 grid where the grid checks its own work so nothing leaves the network). false = skip checking, pay generation only",
+  "type": "boolean"
+}

Changed1 schema field changedSep 2, 2026

addedInput schema / properties / rules

Added value: +{
+  "description": "House rules the assistant must NEVER break, e.g. \"Never promise a refund\". Without them the model invents its own policy, differently each time",
+  "items": {
+    "maxLength": 200,
+    "type": "string"
+  },
+  "maxItems": 12,
+  "type": "array"
+}

AddedAug 31, 2026

TDQS

A4.4/5.0

Behavior5/5

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

With no annotations provided, the description carries the full burden and does so well. It discloses the synchronous response ('returns in seconds... dataset id and a claim token'), the asynchronous follow-up ('generation then runs in the background'), and the critical one-time token behavior with an explicit save warning.

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 short, purposeful sentences. The key usage context, return behavior, and urgent token warning are each front-loaded without fluff or repetition.

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

Completeness4/5

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

Given the complexity of a 14-parameter payment tool with no output schema, the description covers the critical outcomes and the one-time claim token. It doesn't spell out the exact response JSON shape or the next polling step, but those are not essential for invoking the tool correctly in this flow.

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 explains all 14 parameters. The description adds a little value by characterizing paymentPayload as 'signed X-Payment' and emphasizing the quote id, but it does not meaningfully extend the schema's parameter-level documentation.

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 identifies the action as the paid finalization step for buying a training dataset: 'submit the job again with the signed X-Payment payload and the quote id.' It names the concrete outputs (dataset id, claim token) and distinguishes itself from earlier payment-request siblings by being labeled 'Step 2.'

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?

The 'Step 2' framing and the need for a quote id and signed payment payload make the sequential context clear. It does not explicitly name alternatives or say when not to use the tool, but an agent can infer that it comes after request_dataset_payment and before get_dataset_status.

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

All Singularity Layer MCP 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?