MCP For Claude Marketplace
Token Counter and Budget Enforcer

Token Counter and Budget Enforcer MCP Connector for Claude

A+

Estimates token usage and enforces strict cumulative budget limits for AI agents.

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

This MCP server provides precise token estimation and budget management for AI agents. It uses deterministic BPE approximation to calculate token counts for text and code, ensuring predictable resource tracking. The server tracks cumulative usage per agent and per conversation, allowing for strict enforcement of token limits. Use estimate_tokens to predict costs, track_and_enforce_usage to record consumption and check against limits, or get_usage_summary to view current totals. It is ideal for managing LLM costs and preventing budget overruns in multi-agent systems.

tokensbudgetllmusage-trackingestimation

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

estimate_tokens

Calculates the estimated token count for a provided text string

track_and_enforce_usage

Updates the cumulative usage for an agent and a conversation, then checks against set limits

get_usage_summary

Retrieves current cumulative totals without consuming new tokens

See how to talk to your AI agent using Token Counter and Budget Enforcer.

How many tokens will this sentence use: 'Hello, how are you today?'

The estimated token count for that sentence is 6 tokens.

Estimate the tokens for this code: 'print("hello")'

The estimated token count for that code snippet is 4 tokens.

What is the current usage for agent_123?

Agent_123 has consumed a total of 1,250 tokens.

Tokens are estimated using a deterministic character-to-token ratio: approximately 4 characters per token for English text and 3.5 characters per token for code blocks.

Related Connectors

Prompt Length Classifier

Prompt Length Classifier MCP

3 tools Official

Estimate token usage and context window saturation.

A+ View details →
LLM Token Counter

LLM Token Counter MCP

2 tools Official

Calculate exact and estimated token counts for GPT-4, GPT-4o, Claude, and Llama models.

A+ View details →
AI Token Counter

AI Token Counter MCP

1 tools Official

Give AI Agents self-awareness of context limits. Count exact OpenAI/Claude tokens local to prevent fatal API truncation errors.

A+ View details →
Deterministic Timeout Enforcer

Deterministic Timeout Enforcer MCP

3 tools Official

Prevents agent hang-ups by enforcing strict execution time boundaries and token budgets.

A+ View details →