"Optimizing AI Model Thinking, Token Usage, and Context Size" matching MCP servers:

Enables coding agents to retrieve token-budgeted, source-hashed code context from explicitly registered local repositories via read-only MCP tools, reducing broad repository crawling while providing structure, symbols, and impact slices.

Enables agents to calculate how many evaluation test cases are needed to trust model comparisons, find the smallest detectable accuracy gap, and assess paired or unpaired results using McNemar and two-proportion statistical tests.

MCP server for querying OpenAI usage and cost data, including spend summaries, daily breakdowns, month-over-month comparisons, and token usage by model.

Official MCP server for the Wheel Fitment API, giving LLM agents access to vehicle wheel and tire compatibility data, including OEM specs, reverse lookups, plus-sizing calculations, and product card generation.

Simple sequential thinking MCP in python

A MCP server for tracking AI usage metrics and structured logs across applications. Monitor model calls, analyze usage patterns, track costs, and debug AI interactions.

Forces host coding agents through a structured multi-step reasoning protocol, scaffolding, validating, and formatting their thinking in QUICK, STANDARD, or DEEP modes, plus independent refutation and self-consistency verification passes. It runs entirely locally over stdio or Streamable HTTP using bash with no LLM, API key, GPU, or inference cost.

Enables AI agents to monitor live token consumption and rate-limit windows across providers and assess whether expensive actions are justified before exhausting quotas, with optional budget guardrails.

Enables MCP clients to query project-scoped Graphify graphs, search, read, and write Obsidian Markdown notes, and recall or store persistent ICM memories over authenticated private HTTPS. It serializes access per project and supports slim and standard Docker deployments without embeddings or extra memory services.

An MCP server that provides structural codebase indexing and surgical query tools to drastically reduce token usage through symbol-level searches and transitive impact analysis. It supports multiple languages and integrates with git to help AI agents understand code dependencies and the impact of changes in sub-millisecond time.

An MCP server that seats multiple LLMs as a council, letting your assistant query them in parallel or in sequence, relay answers for cross-critique, and merge conclusions within one conversation.

Enables MCP hosts and AI agents to request pay-per-call token risk assessments for contract addresses across Base, Ethereum, and Solana, returning a verdict, 0–100 risk score, six market-structure signals, and an analyst summary, with payment handled via x402 USDC on Base.

MCP server that pings 130+ free coding LLM models across 17 providers in real-time, ranks them by latency, and helps AI agents pick the fastest available model.

Discovers local model runtimes such as Ollama, LM Studio, and LocalAI, and exposes their endpoint metadata to MCP clients like Copilot.

A local-first, multi-provider cost meter for LLM usage, exposed as MCP tools. Captures every call into a local SQLite ledger and lets any coding agent query spend, compare providers, and get recommendations — no cloud, no account. First-class support for Chinese providers (Qwen, DeepSeek) alongside Anthropic and OpenAI.

A fully offline MCP server for token estimation, prompt compression, model routing, and semantic caching to optimize LLM usage costs and efficiency.

MCP server for structured reasoning with cognitive trap detection, verification, and context compression

Facilitates structured, step-by-step problem-solving using the Model Context Protocol, ideal for planning complex tasks like supply chain management.

Manages .ai/ workspace folder, syncs assets with a cloud service, and generates environment-specific bootstrap files for AI context management.

An MCP server that turns the Hugging Face Hub into a conversational model picker with a tool-use eval harness to prove the agent actually uses it correctly.

An advanced MCP server that implements sophisticated sequential thinking using a coordinated team of specialized AI agents (Planner, Researcher, Analyzer, Critic, Synthesizer) to deeply analyze problems and provide high-quality, structured reasoning.

Context window usage estimation for AI coding agents via MCP, enabling proactive state preservation before compaction.

An MCP server that keeps raw tool output out of the context window by sandboxing and compressing results (about 98% reduction), persisting file edits, git operations, tasks and decisions in SQLite with BM25 retrieval so the agent resumes after compaction, and enforcing routing across 17 platforms via MCP and hooks. It also encourages the model to write scripts that compute answers and log only results instead of reading large data into context.

Multi-modal RAG engine for AI assistants. Stores conversation history, conclusions, diffs, error traces, and other development artifacts in LanceDB with vector search, multi-factor scoring, and an LLM-driven consolidation pipeline.

Context Mode is an MCP server that reduces context window waste by sandboxing data-heavy tools, tracking session state in SQLite, and promoting code-based analysis over raw data reads, achieving up to 98% context savings.

An MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programming languages and authenticated CLIs.

Relays a result screenshot to Slack so you can review a coding agent's output without opening the app.

Enables agents to query governed metrics through MCP, including describing metrics, running deterministic SQL against a warehouse, explaining join path preferences, and searching semantically ambiguous terms.

Provides official INDEC register designs and methodological rules to AI models, enabling accurate EPH data analysis code (R/Python) without hallucinations.

Persistent 3D/2D spatial world model MCP server for AI agents. Entity tracking, object permanence, AABB collision simulation & view frustum projection via local SQLite.