标签：#mcp

Fabiane Nardon shares how TOTVS prepares enterprise data for token-hungry AI agents. She discusses balancing deterministic logic and non-deterministic LLMs across precision, security, and cost. Nardon details using data mesh, low-latency database architectures, semantic ontologies, and dynamic MCP tool selection to optimize context windows and reduce token overhead in transactional systems. By Fabiane Nardon

AI agents are becoming first-class citizens in developer workflows. They can read code, run tests, and deploy apps. But one thing they struggle with is understanding database schemas. Database design tools haven't changed in 20 years. You either use a heavyweight desktop app (Navicat, PDManer) or a pretty but closed web app (dbdiagram). Neither supports versioning, real-time collaboration, or AI agent integration. I built ERD Online to solve this. It's an open-source database design tool that combines Git-like versioning with Figma-like collaboration, plus MCP integration for AI agents. In this article, I'll show you how to let Cursor, Claude, or Cline read and write your database schema through MCP, while you keep full control. Database schema changes are hard to track: Who changed what? When did they change it? Why did they change it? How do I rollback? And now with AI agents, there's a new problem: how do you let an AI agent suggest schema changes without giving it a black box that generates random ER diagrams? The wrong approach: ask AI to "generate an ER diagram for an e-commerce app." You get a diagram, but it has no connection to your actual project, no versioning, and no approval flow. The right approach: let the AI agent read your existing schema, suggest changes, and submit them as a version that you review and approve. That's what ERD Online + MCP does. MCP (Model Context Protocol) is a protocol for AI agents to interact with external tools. Think of it as a USB-C port for AI applications. It standardizes how agents discover and call tools. MCP has three main primitives: Tools : Functions the AI can call (like list_projects or create_version ) Resources : Data the AI can read (like project.json ) Prompts : Pre-defined templates for common tasks ERD Online exposes MCP tools that let AI agents: list_projects : List all your ERD projects get_project : Get a project's projectJSON create_version : Suggest a new version of your schema The key boundary: AI agent

An API operation can receive input from several places. Path parameters identify the record. Query parameters filter or paginate the result. Headers carry metadata or authentication. The request body contains structured data for create and update operations. An MCP tool should give the AI client one clear input schema. That is the mapping problem: HTTP API inputs path + query + headers + body become MCP tool input one structured schema the AI client can understand This tutorial walks through that mapping with practical examples. The goal is to make the tool easy for an AI client to call without hiding the real API contract. Example API operation Imagine a project-management API with this endpoint: PATCH /workspaces/{workspace_id}/projects/{project_id}/tasks/{task_id} It updates one task. The API accepts: path parameters for workspace_id , project_id , and task_id ; query parameters such as notify_assignee ; a request body with the fields to update; authentication through a Bearer token header; an optional request header such as Idempotency-Key . A shortened OpenAPI-style version might look like this: paths : /workspaces/{workspace_id}/projects/{project_id}/tasks/{task_id} : patch : operationId : updateTask summary : Update a task description : " Update the title, status, assignee, or due date for one task." parameters : - name : workspace_id in : path required : true schema : type : string - name : project_id in : path required : true schema : type : string - name : task_id in : path required : true schema : type : string - name : notify_assignee in : query required : false schema : type : boolean default : false - name : Idempotency-Key in : header required : false schema : type : string requestBody : required : true content : application/json : schema : type : object properties : title : " " type : string status : type : string enum : [ todo , in_progress , blocked , done ] assignee_id : type : string due_date : type : string format : date minProperties : 1 securi

Caricare un’estensione da disco, aprirne il popup e automatizzare verifiche UI: un workflow più completo per chi sviluppa estensioni e usa agenti. Sviluppare un’estensione Chrome oggi significa spesso alternare tre modalità: codice “a mano”, generazione assistita da un coding agent e una fase di verifica nel browser che resta comunque imprescindibile. Il problema è che molti agenti riescono ad aprire pagine e cliccare elementi, ma si fermano quando entrano in gioco le estensioni: installazione, gestione del popup, interazioni con la UI dell’estensione, verifica rapida dei cambiamenti. Chrome DevTools for agents colma proprio quel vuoto: aggiunge al set di strumenti dell’agente la possibilità di installare e pilotare un’estensione durante i test, oltre a renderne più pratico il debugging. Quando è davvero utile Ci sono alcuni scenari tipici in cui il supporto “estensioni-aware” fa la differenza: Ciclo di feedback più rapido : compili/packi l’estensione, la carichi in Chrome e verifichi subito il popup o una content script UI. Test end-to-end più realistici : invece di simulare una UI in una pagina fittizia, testi l’estensione nel suo contesto reale (action popup, permessi, storage, ecc.). Validazione automatizzata : l’agente può controllare che l’estensione si installi correttamente, che il popup si apra e che i componenti principali siano presenti e interagibili. In pratica: se il tuo agente sa “guidare” il browser ma non sa “gestire” le estensioni, la qualità del test rimane limitata. Setup: abilitare esplicitamente gli strumenti per le estensioni Un dettaglio importante: per ragioni di sicurezza e controllo (in particolare per l’uso dei token e del contesto in cui operano gli agenti), le funzionalità specifiche per estensioni non sono abilitate di default . Dopo aver installato Chrome DevTools for agents, serve quindi un passaggio esplicito nella configurazione MCP: individua il tuo file di configurazione MCP ; abilita la categoria dedicata alle estensioni aggiung

If an AI assistant can write code but cannot reliably hash a value, inspect a JWT, validate JSON, or calculate a CIDR range, you have a small but recurring reliability problem. Asking the model to do those jobs from memory adds an unnecessary interpretation step. DevUtils MCP Server packages 36 everyday developer utilities behind the Model Context Protocol . The server runs locally over standard input and output, so an MCP-compatible client can call explicit tools instead of guessing an operation. This tutorial connects the released 1.1.0 package, verifies the protocol handshake, and shows how to choose a useful tool without treating the server as a replacement for application libraries. TL;DR Install Node.js 18 or newer, add the server command to your MCP client's configuration, restart the client, and ask it to use a tool such as json_validate , jwt_validate , or cidr_calculate . The smallest configuration is a command plus the package name: { "mcpServers" : { "devutils" : { "command" : "npx" , "args" : [ "devutils-mcp-server" ] } } } The released package declares Node.js >=18 . The repository's current default branch has moved ahead to 1.1.1 , so the commands and behavior in this article target the immutable v1.1.0 release and the npm latest package that was verified during research. Prerequisites You need: Node.js 18 or newer and npm. An MCP-compatible client that supports a local stdio server. Permission to run npx and download the public npm package on first use. No API key, account, database, or external service is needed for the local server. The MIT-licensed repository lists Claude Desktop, Cursor, VS Code, Windsurf, Docker, and other MCP-compatible clients as possible consumers. Their configuration file locations differ, but the server entry is the same. Install the released server The release README documents an npx path that does not require a global installation: npx devutils-mcp-server For an automated setup where accepting the package prompt must be e

ToolJet was the first internal tool builder to go AI-native, but ToolJet MCP takes that idea...

MCP is changing how AI applications interact with the systems around them. An AI agent can now do...

Connecting an AI agent to one API is straightforward. Connecting it to many changing APIs—without filling the model context with hundreds of tool definitions—is a different problem. Disclosure: This article was prepared for QVeris and uses QVeris as the implementation example. This tutorial presents a practical pattern for developers building agents that need current external data: discover → inspect → probe → call . Instead of exposing every possible operation up front, the agent discovers the capabilities relevant to the current task, verifies the selected tool, validates its inputs, and only then executes it. TL;DR: Keep the agent's initial tool surface small. Let it discover a capability by intent, inspect the exact schema, probe the request without execution, and make a real call only after the parameters and expected cost are understood. Contents Why a large static tool list becomes difficult The four-step capability workflow Connecting a hosted MCP server A concrete example Production checklist Why a large static tool list becomes difficult An agent connected directly to several providers may need to understand different authentication schemes, parameter conventions, response formats, and error behaviors. Loading every operation into context can also make tool selection less reliable. Model Context Protocol (MCP) provides a standard way for clients to connect to tools and data sources. The protocol solves the connection boundary, but developers still need a strategy for controlling how many capabilities the model sees and when execution is allowed. A compact routing layer is useful when: the agent needs data from multiple API providers; the appropriate provider depends on the user's request; schemas or available operations may change; calls can consume credits or trigger rate limits; you want to validate inputs before executing a paid operation. The four-step capability workflow 1. Discover The agent starts with a natural-language description of the capabilit

Back in July I wrote about what it takes to wire an AI agent into social platforms: six OAuth flows,...

The schema catalog for an AI assistant is the artefact that answers the question "what does this database look like right now". Whether the database is Postgres, MySQL, SQL Server or Redshift, the shape of the problem is the same: the catalog carries table names, column names, types, keys, and enough relationships to let the assistant write a query that resolves. It lives somewhere between the database and the assistant, has to stay in sync with a database that changes underneath it, and is almost always built the same weekend the team decides they want an AI assistant reading their data. It runs fine for the first three tables. The problems start around the fourth week, and none of them look like the same problem twice. The distinction worth naming early is between the connection layer (how the assistant reaches the database) and the knowledge layer (what the assistant knows about the database's shape). The connection layer receives most of the attention, because credentials, network isolation and query cost are visible failure modes and easy to argue about. The knowledge layer is where most of the actual quality of the assistant lives, and it decays quietly. The AI database context page covers why this second layer matters at all when the first one exists. Why not just point the assistant at the database Connecting the AI directly to production is the shortest path and the one most teams reject after five minutes of thinking about it. The assistant would get read access on tables it should not see, its queries can be arbitrarily expensive, its credentials would live somewhere they should not, and the audit trail becomes hard to reason about. What most teams end up building is a layer in between: a representation of the database that the assistant can read cheaply and safely without ever touching production. That layer is what this article is about. It is not the connection. It is the catalog. The five recipes teams build Ask fifteen senior developers how to build

First post in a build-in-public series about verdict , an MCP server that gives coding agents structured, sandboxed test feedback. The problem Watch a coding agent work and you'll see it run pytest in your shell, unsandboxed, and then push 40,000 tokens of raw output through its context window to answer one question: did my change break anything? That's three problems in one command: Token waste. The agent needs ~10 lines of signal and pays for a wall of dots, warnings, and tracebacks. No sandbox. The tests run on your machine, in your environment, with your files writable. No memory. When a test fails, the agent can't tell whether it broke it or whether it was broken before it arrived - so it either "fixes" pre-existing failures nobody asked about, or ships regressions it assumes were already there. verdict is an MCP server that replaces the pytest shell-out with four tools: tool what it returns verify(scope?) impact-selected tests, run in an ephemeral container, as a ~400-token typed verdict explain_failure(check_id) the full traceback - only on demand history(fingerprint) first seen / last seen / times seen for a failure run_checks(["ruff","mypy"]) lint & type checks, same verdict shape ▶️ Watch the 30-second demo - Claude Code fixing a bug with verdict verifying in a container. The three ideas 1. Verdicts, not output. verify returns typed JSON: counts, per-failure message + location, and nothing else. Full tracebacks live behind explain_failure . The whole verdict for a real failing run is ~400 tokens - the raw pytest output it replaces was ~40k. The design rule in the repo is blunt: nothing bulky rides in the summary, ever. 2. Fingerprints give failures identity. Every failure is hashed from its normalized signature - volatile tokens (addresses, tmp paths, ids, durations) collapsed first. Same logical failure ⇒ same fingerprint, across runs and refactors. Fingerprints are what make the third idea possible: 3. History answers "was it me?" verdict keeps a small S

If you're wiring MCP servers into an agent and want to check them for prompt injection, tool poisoning, or supply-chain risk before you trust them, there are now a handful of open-source options. This is a factual, no-benchmarks comparison of the three I could actually find and read the docs for: our own sentinel-scan-cli , Cisco's mcp-scanner , and what used to be Invariant Labs' mcp-scan . One thing worth flagging up front: Invariant Labs' mcp-scan repo ( github.com/invariantlabs-ai/mcp-scan ) now redirects to github.com/snyk/agent-scan . The project has been absorbed into Snyk and rebranded as "Agent Scan" (package snyk-agent-scan ). If you're comparing tools based on older blog posts that reference "Invariant Labs mcp-scan" as a standalone, no-account CLI, that's out of date — running it now requires a free Snyk account and an SNYK_TOKEN API key ( export SNYK_TOKEN=... ) before the CLI will scan anything. I'm comparing against the current Snyk Agent Scan README since that's what the repo actually ships today. All claims below are pulled directly from each project's public README as of 2026-08-24. No invented features, no synthetic benchmarks — this is a "what does the doc actually say" comparison, not a lab test. Feature comparison sentinel-scan-cli Cisco mcp-scanner Snyk Agent Scan (fka Invariant Labs mcp-scan) License MIT Apache 2.0 source-available on GitHub; requires Snyk account/token to run Install zero dependencies, single Python file or pip install / npx github:... uv tool install , Python 3.11+ uvx snyk-agent-scan or standalone binary Signup / API key required to run at all No ( --demo needs nothing; scanning your own endpoint needs only your own endpoint's key) No (core YARA/static scanning works with zero keys; LLM/Cisco AI Defense/VirusTotal analyzers are opt-in extras) Yes — Snyk account + SNYK_TOKEN required before any scan runs What it scans Live LLM endpoint (prompt-injection/jailbreak suite) and static MCP tool manifests ( mcp.json ) Live MCP se

When you run multiple agents across Claude, ChatGPT, and Cursor, each one starts from scratch unless you manually paste context into every session. OzBrain solves this by exposing a shared knowledge substrate that agents read and write through the Model Context Protocol (MCP). The system routes context so agents see only what they need, and teams avoid explaining the same facts to every new agent instance. The Show HN post drew 85 points and 50 comments because the problem is real: production multi-agent workflows break down when context lives in isolated chat histories or scattered documents. OzBrain's architecture treats knowledge as a first-class resource with explicit scoping, indexing, and conflict resolution. Storage Layer and Scope Boundaries OzBrain organizes knowledge into brains , which are either personal or shared. Each brain holds structured knowledge units that agents query through the MCP connector. The system decides scope at write time: Personal brains store user-specific preferences, writing style, and private project state. Shared brains hold team-wide facts like client contacts, project decisions, and open threads. When an agent writes to OzBrain, it specifies the target brain. The MCP connector enforces access control: agents can read from any brain the user has joined, but write permissions depend on the brain's sharing policy. This prevents accidental leakage of personal context into team memory. The storage layer tags each knowledge unit with metadata: creation timestamp, last update, and a freshness indicator (fresh, aging, stale). Agents use these tags to decide whether to trust the stored fact or re-query the source. Indexing Strategy and Query Routing OzBrain does not load the entire knowledge graph into every prompt. Instead, it maintains a routing index that maps topics to knowledge units. When an agent queries for "client contacts," the index returns pointers to relevant units without pulling in unrelated project state. The routing ind

Originally published on michael-kaminski.io . The Genome of Games publishes the same 1,180 records four different ways, and one command writes all four: node build.js , 0.39 seconds, zero npm dependencies. Out come 1,245 static HTML pages for crawlers, an interactive canvas graph for humans, a 129,037-byte search index for the site's own search box, and a Model Context Protocol server exposing 8 tools to agents. The decision worth copying is the one that sounds like a downgrade. The MCP server does not query the site and does not read the source data. It statically imports a 1.9 MB index that the build wrote. There is exactly one place where slugs, lineage, and adoption edges get joined, so an agent and a crawler cannot come back with different answers. The dataset is an ontology of video game mechanics — 168 mechanics, 618 games, 394 companies, 4,366 recorded links, 1962 to 2025. What the records are about does not matter here. The shape of the problem shows up anywhere a structured knowledge base has to serve both a search engine and a model. Four surfaces, one build, a twelve-fold expansion Six hand-edited JSON files under data/ are the source of truth: the feature ontology, the graph, the prose, the company registry, the site copy, and the verified outbound links. Together they are 1,312,577 bytes. The build turns that into 16,644,215 bytes of generated read surface. A 12.7× expansion, and every byte of it is disposable. Surface Consumer Bytes Per entity 1,245 static HTML pages Crawlers, humans 14,613,203 11,728 / page mcp-index.json → MCP server Agents 1,901,975 1,612 search-index.json The site's own search box 129,037 109 /graph/ canvas Humans exploring lineage data injected at build — The build also emits sitemap.xml with 1,245 entries, llms.txt , robots.txt , and a 404 page. The same run reports 96,843 internal links across those pages. Nothing in that list is authored. Delete the whole output directory and the next build restores it in under half a second.

If you invoice or onboard Polish companies, sooner or later you have to check two dull things that turn out to matter a lot: is this company actually a registered VAT payer, and is the bank account they gave you the one that's on the government's official white list ("Biała Lista")? Both of those affect whether you can deduct the cost and reclaim VAT, so it's not really optional. The annoying part is that the data lives in four different places: the Ministry of Finance, the KRS court register, GUS (the stats office), and the EU's VIES service. Each one has its own API and its own quirks. I got tired of gluing those together every time, so I wrapped them behind a few plain HTTP calls that return JSON. Full disclosure: skanfirmy.pl is mine. It's free, no key, no signup, and the web layer runs client-side with no tracking. Here's how you'd actually use it. REST: one GET, one JSON Cheapest thing you can do is check a NIP (the tax ID): curl https://skanfirmy.pl/nip/5260250995 You get back the VAT status (active, exempt, or not registered), the company details from the VAT register, and the accounts sitting on the white list. The paths: GET /nip/{nip} gives VAT status + white-list data for one NIP GET /nips/{list} takes several NIPs at once (comma-separated) GET /regon/{nip} returns data from the REGON register (GUS) GET /vies/{country}/{number} validates an EU VAT number, e.g. /vies/DE/811128135 It's a plain GET that returns JSON, so it drops into anything that can make an HTTP request: a cron job, a lambda, a CI step, whatever. Python requests and a few lines. This one raises if the company isn't an active VAT payer: import requests def check_vat ( nip : str ) -> dict : r = requests . get ( f " https://skanfirmy.pl/nip/ { nip } " , timeout = 10 ) r . raise_for_status () data = r . json () status = data . get ( " vatStatus " ) or data . get ( " status " ) if status != " Czynny " : # status comes back in Polish; compare against the raw value raise ValueError ( f " NIP { n

A podcast host recently told me that he prepares questions from the guest’s bio using ChatGPT. That works for the basics, but a bio does not show which stories the guest has repeated across other interviews or which questions they have already answered many times. I’m helping Audiogram test a different workflow. It connects to Claude through MCP, searches Apple Podcasts, retrieves available episode transcripts, and lets Claude compare the guest’s previous answers before drafting new questions. For one test, I used two published Sam Altman interviews. The workflow pulled both available transcripts, separated recurring themes from open gaps, and produced follow-up questions around measurable evidence, privacy limits, and independent review—rather than repeating another general “will AI be good or bad?” question. The prompt is simple: Prepare an interview brief for [guest] about [interview angle]. Find podcast episodes where the guest is actually interviewed, retrieve the available transcripts, and compare them. Show recurring themes, changes in position, questions already answered, and five follow-up questions based on gaps or unsupported claims. Cite the podcast and episode for every finding. Separate transcript evidence from inference, and say what is missing when the available material is not enough. This is for research across published Apple Podcasts episodes. It is not a raw-audio editor, and transcript availability and speaker labels still need to be checked. You can see the complete recipe and tested example here: Podcast guest interview preparation with Audiogram If you prepare podcast interviews, would previous-interview comparison improve your questions, or is another part of guest research still the bigger problem? Disclosure: I’m helping Audiogram with early-user growth and used AI to help edit this post.

MCP Was a Mistake. Here Are 200,000 Tokens That Prove It. "mcp were a mistake. bash is better." — Peter Steinberger, OpenClaw founder I didn't want to believe it either. MCP was supposed to be the USB-C of AI — one protocol to connect everything. Anthropic, OpenAI, Google all backed it. 97 million monthly downloads. 17,000 servers. But then I measured what MCP actually does to your context window. The Setup I connected 10 popular MCP servers to a token counter. Here's what happened before I typed a single word: Server Tools Tokens Injected Filesystem 11 3,847 Brave Search 8 2,103 Sequential Thinking 3 890 Memory 9 2,567 Puppeteer 15 5,890 Postgres 19 8,231 Notion 24 13,780 GitHub 28 12,440 Slack 22 14,672 Google Drive 31 47,293 Total 170 111,713 111,713 tokens. Before your first message. That's not a typo. Connecting 10 MCP servers to Claude means over 100K tokens of JSON schemas get injected into your context window. You haven't asked a question yet. You haven't made a tool call. The schemas are just... sitting there. The Math That Made Me Angry At Claude 3.5 Sonnet pricing ($3/M input tokens): Every conversation starts with 111K tokens of overhead: $0.33 20 conversations per day: $6.67/day 22 working days per month: $147/month Annual cost of JSON schemas: $1,764 That's more than a Claude Pro subscription. You're paying $1,764/year to read JSON braces describing tools you might never use. But Wait — It Gets Worse The 111K is just the schema injection. When you actually call a tool, MCP wraps the result: { "content" : [ { "type" : "text" , "text" : "{ \" file \" : \" app.py \" , \" size \" : 1024}" } ] } The actual content is 38 characters. The wrapping is 47 characters. 55% of your result tokens are JSON overhead. With 20 tool calls per conversation: Schema injection: ~111K tokens Result wrapping: ~18K tokens Total overhead: ~130K tokens per conversation Your $0.54 conversation now has 130K tokens that serve zero purpose. What Garry Tan Was Right About When YC's CE

Claude Code Is Burning Your Token Budget. Here's the Receipt. I found $2,500/year of hidden token waste in my Claude Code setup. It was the MCP servers. The Discovery Last week I noticed my Claude Code conversations were dying at around message 15. Context window full. The model starts forgetting earlier instructions. Tool calls fail. The conversation degrades into hallucination. I assumed it was my fault — too many messages, too much context. So I started measuring. Here's what I found: Session start: Claude system prompt: ~8,000 tokens MCP schema injection: ~111,000 tokens User's first message: 50 tokens ────────────────────────────────────────── Total before any work: ~119,000 tokens Remaining context: ~81,000 tokens I was starting every conversation with 60% of my context already consumed. The culprit wasn't my prompts. It was the 10 MCP servers I had proudly configured in my claude_desktop_config.json . The Receipts I measured each server's schema injection using tiktoken: Server Why I Installed It Token Cost Times Used/Week GitHub PR reviews, issues 12,440 3 Slack Message reading 14,672 0 Google Drive Doc access 47,293 1 Notion Knowledge base 13,780 2 Postgres Query DB 8,231 4 Puppeteer Screenshots 5,890 0 Filesystem File access 3,847 15 Brave Search Web search 2,103 5 Memory Context persistence 2,567 0 Sequential Thinking Reasoning 890 2 Total 111,713 Look at the "Times Used/Week" column. Three servers were used zero times. Two more were used once or twice. But every single one of them was injecting 100% of its schema into every conversation. I was paying $0.33 per conversation — $2,500/year — to load schemas for tools I barely used. The Moment I Realized Everyone Has This Problem I posted my findings on Bluesky. Within hours: "I had the same issue. Removed 6 MCP servers and my conversations went from dying at message 15 to lasting 40+ messages." — @developer1 "GitHub MCP is 12K tokens but Claude Code already has gh CLI built in. Why did I install it?" — @dev

Garry Tan Was Right: "MCP Sucks Honestly." I Have the Token Receipts. "MCP sucks honestly. Context window eats too much, auth is a mess. I wrote a CLI wrapper in 30 minutes and it works better." When YC's CEO says this on X, people listen. But nobody had the data to back it up. Until now. What Garry Tan, Perplexity's CTO, and 97 Million Downloads Can't Hide Three things happened in the last 6 months that changed how I think about MCP: Peter Steinberger (OpenClaw founder) tweeted: "mcp were a mistake. bash is better." Eric Holmes wrote "MCP is dead. Long live the CLI" — it hit HN frontpage Denis Yarats (Perplexity CTO) publicly announced they're replacing MCP with REST API + CLI internally Garry Tan (YC CEO) replied: "MCP sucks honestly" The community split into two camps: "MCP is dead" — CLI is simpler, cheaper, faster "MCP is fine" — 97M downloads, 17K servers, it's the standard Both are wrong. The problem isn't MCP. The problem is what MCP does to your context window. The 47,000-Token Problem Nobody Measured I connected 10 MCP servers to a token counter. Here's what I found: MCP Server Tools Token Cost Equivalent Sequential Thinking 3 890 This blog post Brave Search 8 2,103 A short email Filesystem 11 3,847 A README Memory 9 2,567 A meeting note Puppeteer 15 5,890 A chapter of a book Postgres 19 8,231 A whitepaper GitHub 28 12,440 A court filing Notion 24 13,780 A legal contract Slack 22 14,672 A novella chapter Google Drive 31 47,293 Half of a novel Total 170 111,713 A short book One MCP server — Google Drive — injects 47,293 tokens into your context before you ask a single question. The entire works of Shakespeare is 900K tokens. Google Drive's schema is 5% of Shakespeare. For listing files. The Cost Breakdown (So You Can Get Angry Too) At Claude 3.5 Sonnet pricing ($3/M input tokens, $15/M output): Scenario Tokens Cost Annual Cost 1 server (minimal) 3,847 $0.01/conv $4.40/yr 3 servers (common) 14,528 $0.04/conv $19.40/yr 5 servers (typical) 33,061 $0.10/conv $4

I Benchmarked 10 MCP Servers — One of Them Burns 47K Tokens Just to Say Hello 10 popular MCP servers. 847 tools total. 312K tokens of JSON schemas. One server alone wastes more tokens than a full GPT-3 conversation. Here are the results. What I did I installed the 10 most popular MCP servers from the official registry. Connected each one to a token counter. Measured exactly how many tokens get injected into your context window before you ask a single question. The servers: # Server Tools Token Cost 1 Filesystem 11 3,847 2 GitHub 28 12,440 3 Postgres 19 8,231 4 Puppeteer 15 5,890 5 Brave Search 8 2,103 6 Memory 9 2,567 7 Sequential Thinking 3 890 8 Slack 22 14,672 9 Google Drive 31 47,293 10 Notion 24 13,780 Totals: 847 tools across 10 servers 111,713 tokens of JSON schemas 200,000+ tokens including server status messages, headers, and error schemas That's right — connecting 10 MCP servers to Claude means 200K tokens of overhead before your first message . The worst offender: Google Drive Google Drive's MCP server exposes 31 tools. Each tool has deeply nested schemas for file operations, permission management, sharing, and search. The full schema dump: { "name" : "drive.files.list" , "description" : "Lists files in the user's Google Drive with optional filtering" , "inputSchema" : { "type" : "object" , "properties" : { "q" : { "type" : "string" , "description" : "Query string for filtering files..." }, "corpora" : { "type" : "string" , "enum" : [ "user" , "domain" , "sharedDrive" , "allDrives" ]}, "includeItemsFromAllDrives" : { "type" : "boolean" }, "orderBy" : { "type" : "string" }, "pageSize" : { "type" : "integer" }, "pageToken" : { "type" : "string" }, "spaces" : { "type" : "array" , "items" : { "type" : "string" }}, "supportsAllDrives" : { "type" : "boolean" }, "fields" : { "type" : "string" } }, "required" : [] } } That's ONE tool. 31 of them. At ~1,525 tokens per tool average. 47,293 tokens. Just for Google Drive. For comparison, the entire works of Shakespea