Skip to main content
Glama

RationalBloks

by io.github.rationalbloks
OverviewDiscussionsAdmin

Create Project

create_project

Create a new RationalBloks project from a JSON schema.

⚠️ CRITICAL RULES - READ BEFORE CREATING SCHEMA:

  1. FLAT FORMAT (REQUIRED): ✅ CORRECT: {users: {email: {type: "string", max_length: 255}}} ❌ WRONG: {users: {fields: {email: {type: "string"}}}} DO NOT nest under 'fields' key!

  2. FIELD TYPE REQUIREMENTS: • string: MUST have "max_length" (e.g., max_length: 255) • decimal: MUST have "precision" and "scale" (e.g., precision: 10, scale: 2) • datetime: Use "datetime" NOT "timestamp" • ALL fields: MUST have "type" property

  3. AUTOMATIC FIELDS (DON'T define): • id (uuid, primary key) • created_at (datetime) • updated_at (datetime)

  4. USER AUTHENTICATION: ❌ NEVER create "users", "customers", "employees" tables with email/password ✅ USE built-in app_users table

    Example: { "employee_profiles": { "user_id": {type: "uuid", foreign_key: "app_users.id", required: true}, "department": {type: "string", max_length: 100} } }

  5. AUTHORIZATION: Add user_id → app_users.id to enable "only see your own data"

    Example: { "orders": { "user_id": {type: "uuid", foreign_key: "app_users.id"}, "total": {type: "decimal", precision: 10, scale: 2} } }

  6. FIELD OPTIONS: • required: true/false • unique: true/false • default: any value • enum: ["val1", "val2"] • foreign_key: "table.id"

AVAILABLE TYPES: string, text, integer, decimal, boolean, uuid, date, datetime, json, uuid_array, integer_array, text_array, float_array

Array types store PostgreSQL native arrays with automatic GIN indexing: • uuid_array: UUID[] — for sets of references (e.g., tensor coordinates) • integer_array: BIGINT[] — for dimension indices, integer sets • text_array: TEXT[] — for tags, categories, label sets • float_array: DOUBLE PRECISION[] — for weight vectors, scores GIN-indexed operators: @> (contains), <@ (contained_by), && (overlaps)

BACKEND ENGINE: • python (default): FastAPI backend — mature, full-featured • rust: Axum backend — faster cold starts, lower memory, high performance

WORKFLOW:

  1. Use get_template_schemas FIRST to see valid examples

  2. Create schema following ALL rules above

  3. Call this tool (optionally choose backend_type: "python" or "rust")

  4. Monitor with get_job_status (2-5 min deployment)

After creation, use get_job_status with returned job_id to monitor deployment. While RationalBloks is being updated, the call is refused with 'RationalBloks is being updated': call it again in a few minutes.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
nameYesProject name
schemaYesJSON schema in FLAT format (table_name → field_name → properties). Every field MUST have a 'type' property. Use get_template_schemas to see valid examples.
cluster_idYesREQUIRED — BYOC resource pool ID (from list_clusters) to deploy this project onto your own cluster. Owned hosting is retired: a project we operate must run on your own infrastructure. Register a pool via the Resource Pools UI first, then pass its id here.
backend_typeNoBackend engine: 'python' (FastAPI, default) or 'rust' (Axum, faster). Default: python

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • removedInput schema / properties / description
      Removed value: -{
-  "description": "Optional project description",
-  "type": "string"
-}
  2. Changed2 schema fields changed
    • changedInput schema / properties / cluster_id / description
      Previous value: -"Optional BYOC resource pool ID (from list_clusters) to deploy this project onto your own cluster. Omit to deploy on RationalBloks-hosted infrastructure."New value: +"REQUIRED — BYOC resource pool ID (from list_clusters) to deploy this project onto your own cluster. Owned hosting is retired: a project we operate must run on your own infrastructure. Register a pool via the Resource Pools UI first, then pass its id here."
    • changedInput schema / required
      Previous value: -[
-  "name",
-  "schema"
-]New value: +[
+  "name",
+  "schema",
+  "cluster_id"
+]
  3. Changed1 schema field changed
    • addedInput schema / properties / cluster_id
      Added value: +{
+  "description": "Optional BYOC resource pool ID (from list_clusters) to deploy this project onto your own cluster. Omit to deploy on RationalBloks-hosted infrastructure.",
+  "type": "string"
+}
  4. Changed1 schema field changed
    • addedInput schema / properties / backend_type
      Added value: +{
+  "description": "Backend engine: 'python' (FastAPI, default) or 'rust' (Axum, faster). Default: python",
+  "enum": [
+    "python",
+    "rust"
+  ],
+  "type": "string"
+}
  5. First observed

TDQS

A4.7/5.0
Behavior5/5

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

The description discloses that creation is asynchronous via a returned job_id, that deployment takes 2-5 minutes, and that calls may be refused with a specific message while updates occur. It also documents strict schema validation requirements. These behavioral details go well beyond the annotations, which only state readOnly=false, openWorld=true, idempotent=false, and destructive=false.

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?

The description is long but exceptionally well structured with headings, numbered rules, examples, and a workflow. It is front-loaded with the purpose and then organized by critical constraints. Some repetition exists across examples, but each block serves a distinct instructional function, making the length justified.

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, the description properly supplies the return/job monitoring path and expected deployment duration. It also covers prerequisites, cluster_id requirements, backend selection, schema validation rules, and retry behavior. An agent has enough context to invoke the tool correctly and recover from failures.

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

Parameters5/5

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

Although the input schema covers all 4 parameters (100% coverage), the description adds substantial meaning: flat-format rules, per-type requirements like max_length and precision/scale, automatic fields to omit, authentication/authorization guidance, array type semantics with GIN indexing, and backend engine tradeoffs. This is exactly the practical knowledge an agent needs beyond the schema.

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 opens with a precise verb and resource: 'Create a new RationalBloks project from a JSON schema.' This clearly identifies the tool's core function and distinguishes it from graph/schema siblings like create_graph_project. The extensive schema rules reinforce the exact purpose without ambiguity.

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 WORKFLOW section gives explicit sequencing: use get_template_schemas first, then call this tool, then monitor with get_job_status. It also states retry behavior when RationalBloks is updating. However, it does not explicitly exclude alternatives such as 'use create_graph_project for graph projects,' so it stops short of a 5.

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

All RationalBloks toolsInspect
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