MCP For Claude Marketplace
JSON Schema Function Call Validator

JSON Schema Function Call Validator MCP Connector for Claude

A+

Validates raw LLM text outputs against JSON Schemas to detect missing parameters, type mismismatches, and hallucinations.

3 tools Official Updated Oct 1, 2026 Official Vinkius Partner
More Details Connect to Claude

This MCP server provides essential tools for LLM orchestration and reliability. It allows agents to parse messy text outputs and validate them against strict JSON Schemas. Use validate_function_call to identify missing required parameters, type mismatches, or hallucinated keys. The extract_json_from_text tool helps isolate JSON blocks from conversational filler, while get_validation_summary provides high-level reliability metrics across multiple attempts. This is critical for ensuring that AI agents follow structured function-calling protocols accurately.

json-schemallmvalidationfunction-callingdata-integrity

3 tools expose this connector's capabilities to your AI agent.

extract_json_from_text

A utility to isolate the JSON component from a messy LLM text string

get_validation_summary

Provides a high-level overview of multiple validation attempts to track model reliability over time

validate_function_call

Validates a raw LLM text string against a specific JSON Schema to identify structural and logical errors

See how to talk to your AI agent using JSON Schema Function Call Validator.

Validate this LLM output: 'Here is the call: {"action": "get_weather", "city": 123}' against a schema where city must be a string.

{"missingCount": 0, "typeMismatchCount": 1, "hallucinationCount": 0, "validityPercentage": 50.0, "errorLog": ["Parameter 'city' expected type string, but received integer."]}

Extract the JSON from this text: 'Sure! Here is your data: {"id": 1, "name": "test"} hope this helps!'

{"id": 1, "name": "test"}

Check if this function call is valid: {"user_id": "abc-123"} using a schema that requires 'user_id' and 'session_id'.

{"missingCount": 1, "typeMismatchCount": 0, "hallucinationCount": 0, "validityPercentage": 50.0, "errorLog": ["Missing required parameter: session_id"]}

The `extract_json_from_text` tool specifically isolates the JSON block by finding the first and last curly braces, allowing the validator to ignore surrounding markdown or conversational filler.

Related Connectors

Langfuse (LLM Tracing & Evals)

Langfuse (LLM Tracing & Evals) MCP

10 tools Official

Monitor LLM apps via Langfuse — track traces, manage prompt templates, and audit evaluation scores.

A+ View details →
Char-to-Token Estimator

Char-to-Token Estimator MCP

3 tools Official

Predict token usage for different LLMs using language-specific character ratios.

A+ View details →
TOML Strict Config Parser

TOML Strict Config Parser MCP

3 tools Official

A deterministic, dependency-free parser that transforms TOML strings into structured JSON objects.

A+ View details →
Token Count Estimator

Token Count Estimator MCP

3 tools Official

Deterministic LLM token estimation using character-based heuristics.

A+ View details →