Skip to main content

Documentation Index

Fetch the complete documentation index at: https://spec.conveniencepro.cc/llms.txt

Use this file to discover all available pages before exploring further.

​
Validation

CTP provides built-in validation for tool definitions, inputs, and outputs.

​
Definition Validation

Validate tool definitions against the CTP schema: 
import { validateToolDefinition } from '@conveniencepro/ctp-core';

const result = validateToolDefinition(myDefinition);

if (!result.valid) {
  result.errors.forEach(error => {
    console.error(`${error.path}: ${error.message}`);
  });
}

​
Validation Result

interface ValidationResult {
  valid: boolean;
  errors: ValidationError[];
}

interface ValidationError {
  path: string;      // JSON path to error (e.g., "parameters[0].name")
  message: string;   // Human-readable error message
  code: string;      // Error code (e.g., "MISSING_REQUIRED")
}

​
Common Validation Errors

CodeMessageFix
MISSING_REQUIREDRequired field missingAdd the required field
INVALID_IDID must be lowercase hyphen-separatedUse format my-tool-name
INVALID_CATEGORYUnknown categoryUse one of 8 valid categories
DUPLICATE_PARAMETERParameter name already usedUse unique parameter names

​
Input Validation

Validate parameters before execution: 
import { validateParameters } from '@conveniencepro/ctp-core';

const result = validateParameters(myDefinition.parameters, userInput);

if (!result.valid) {
  return {
    success: false,
    error: result.errors[0].message,
    errorCode: 'INVALID_INPUT',
  };
}

​
Automatic Runtime Validation

The runtime can validate automatically: 
import { createRuntime } from '@conveniencepro/ctp-runtime';

const runtime = createRuntime(registry, {
  validateInput: true,   // Validate before execution
  validateOutput: true,  // Validate after execution
});

// Validation errors returned automatically
const result = await runtime.execute('my-tool', { /* invalid params */ });
// result.errorCode === 'INVALID_INPUT'

​
Parameter Constraints

​
String Validation

{
  name: 'username',
  type: 'text',
  validation: {
    minLength: 3,
    maxLength: 20,
    pattern: '^[a-z0-9_]+$',
  },
}

​
Number Validation

{
  name: 'quantity',
  type: 'number',
  validation: {
    min: 1,
    max: 100,
    step: 1,
  },
}

​
File Validation

{
  name: 'document',
  type: 'file',
  validation: {
    accept: ['.json', '.txt', 'application/json'],
    maxSize: 5242880,  // 5MB
  },
}

​
Manual Validation in Tools

Implement validation in your tool function: 
export const myFn: ToolFunction = (params) => {
  const input = params.input;

  // 1. Required check
  if (input === undefined || input === null || input === '') {
    return {
      success: false,
      error: 'Input is required',
      errorCode: 'MISSING_REQUIRED',
    };
  }

  // 2. Type check
  if (typeof input !== 'string') {
    return {
      success: false,
      error: 'Input must be a string',
      errorCode: 'TYPE_ERROR',
    };
  }

  // 3. Length check
  if (input.length < 1) {
    return {
      success: false,
      error: 'Input must not be empty',
      errorCode: 'CONSTRAINT_VIOLATION',
    };
  }

  if (input.length > 100000) {
    return {
      success: false,
      error: 'Input exceeds maximum length of 100,000 characters',
      errorCode: 'CONSTRAINT_VIOLATION',
    };
  }

  // 4. Pattern check
  if (!/^[\s\S]*$/.test(input)) {
    return {
      success: false,
      error: 'Input contains invalid characters',
      errorCode: 'INVALID_INPUT',
    };
  }

  // Validation passed, proceed with processing
  return {
    success: true,
    data: process(input),
  };
};

​
Result Validation

Validate tool output: 
import { validateToolResult } from '@conveniencepro/ctp-core';

const result = myFn(params);
const validation = validateToolResult(result);

if (!validation.valid) {
  console.error('Tool returned invalid result:', validation.errors);
}

​
Result Requirements

// ✅ Valid success result
{ success: true, data: { /* ... */ } }

// ✅ Valid error result
{ success: false, error: 'message', errorCode: 'INVALID_INPUT' }

// ❌ Invalid: missing success
{ data: { /* ... */ } }

// ❌ Invalid: success true but no data
{ success: true }

// ❌ Invalid: success false but no error
{ success: false }

​
Validation Helper Functions

import {
  isValidToolId,
  isValidCategory,
  isValidParameterType,
} from '@conveniencepro/ctp-core';

// Check ID format
isValidToolId('my-tool');        // true
isValidToolId('My Tool');        // false
isValidToolId('my_tool');        // false

// Check category
isValidCategory('formatters');   // true
isValidCategory('custom');       // false

// Check parameter type
isValidParameterType('text');    // true
isValidParameterType('string');  // false

​
Schema Validation

Use JSON Schema for advanced validation: 
import Ajv from 'ajv';
import toolDefinitionSchema from '@conveniencepro/ctp-spec/schemas/tool-definition.schema.json';

const ajv = new Ajv();
const validate = ajv.compile(toolDefinitionSchema);

const valid = validate(myDefinition);
if (!valid) {
  console.error(validate.errors);
}

​
Best Practices

Check required parameters and basic types before doing any processing.
Error messages should tell users exactly what’s wrong and how to fix it.
Include suggestion in error results to help users fix issues.
{
  success: false,
  error: 'Invalid JSON syntax',
  errorCode: 'INVALID_INPUT',
  suggestion: 'Check for missing commas or unquoted keys',
}
Enforce min/max lengths, ranges, and patterns to prevent edge cases.