iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
No universal MCP rule caps a server at 90 tools. The MCP tools specification supports paginated discovery, so a catalog problem above that size is more likely to involve a particular client’s pagination or aggregation behavior, prompt/context pressure, or the model’s ability to select the right tool—not an established protocol ceiling. The key is to identify which layer fails.
Does MCP limit a server to 90 tools?
No. The MCP specification’s tools/list operation supports pagination and does not set a maximum of 90 tools. A server that declares the tools capability must return the tools currently available to the requesting client. The list may vary with authorization, but the specification says it must not vary per connection or as a side effect of other requests on that connection. See the MCP Tools specification.
That protocol-level behavior does not guarantee that every host or SDK will retrieve, retain, display, or use an arbitrarily large catalog successfully. “90+ tools” is therefore not a general MCP breaking point. It can describe a failure in a particular implementation, but the host, SDK, version, and observed failure need to be identified before assigning a threshold.
Free tools Windows power users keep installed
One-click scans. No signup required.
Where a large tool catalog can fail
Tool discovery involves distinct layers. A failure at one does not show that the MCP server itself has reached a protocol limit.
#1 Best Overall
- 【High-Performance APU】The MS-S1 MAX features an AMD Ryzen AI Max+ 395 APU, integrating a Zen 5 architecture CPU (up to 5.1GHz, 16C/32T, 64M L3 Cache), an RDNA 3.5 GPU, and an NPU (50 TOPS). The total system output is 126 TOPS. It provides powerful parallel computing capabilities for demanding AI workflows. It is ideal for running local LLMs, multimodal models, and computationally intensive tasks
- 【128GB UMA Memory】Equipped with up to 128GB of LPDDR5x-8000MT/s unified memory, it enables the CPU and GPU to access a shared, high-bandwidth memory pool with extremely low latency. Ideal for large-scale AI inference, 3D workloads, and complex timelines in video editing. It eliminates traditional VRAM bottlenecks, ensuring smoother data transfer during high-intensity computations. The UMA design maximizes performance stability under high loads
- 【Flexible Expansion】The MS-S1 MAX features USB4 V2 (up to 80Gbps), dual 10GbE LAN, HDMI 2.1 (up to 8K60), a full-length PCIe x16 expansion slot, and dual M.2 slots supporting up to 16TB RAID 0/1. Wi-Fi 7 provides stronger signal coverage and a more stable wireless experience. The slide-out design facilitates upgrades and maintenance. It easily adapts to personal, studio, or rack-mount enterprise environments
- 【High-Efficiency Cooling System】Utilizing an aerospace-grade aluminum alloy chassis, copper base plate, six heat pipes, dual turbine fans, and advanced PCM thermal conductive material, it maintains stable cooling performance even under continuous load. This system supports 130W continuous power and 160W peak power operation, with a built-in 320W power supply. It boasts multiple global certifications including CCC, FCC, UL, CE, and UKCA, ensuring stable and reliable operation in various environments
- 【Cluster Design】Two MS-S1 MAX units can be configured as a dual-unit cluster to run a large 235B Q4 model locally, achieving an output speed of 10.87 tok/s. Supporting 2U rack deployment, multiple MS-S1 MAX units can be cascaded into a distributed cluster to create a high-efficiency AI computing center. A cluster of four MS-S1 MAX units successfully ran a DeepSeek-R1 671B Q4 large model. A reserved cluster power-on interface allows for unified start-up and shutdown
- Server response: The server may fail to return a valid page or may not implement pagination correctly. This is a server or protocol-implementation issue, not a standard 90-tool cap.
- Client pagination or aggregation: A client may stop following pages, impose a page-count safeguard, or fail while combining results. This can leave tools undiscovered even though the server supports pagination.
- Prompt and context pressure: A host may pass many tool names and descriptions to a model at once. A large catalog can consume context and make the relevant choices harder to distinguish.
- Tool selection: The tools may be available to the model, but it may choose an unsuitable one or miss the right one among similar candidates. That is a selection-quality issue, not a listing failure.
What the TypeScript SDK’s 64-page default means
The MCP TypeScript SDK documents that listTools() called without a cursor automatically traverses pages and aggregates the results. That automatic path has a default listMaxPages limit of 64 pages. Calling with an explicit cursor fetches an individual page without that automatic aggregation cap. See the MCP TypeScript SDK documentation.
64 is a page count, not a tool count. The cited documentation does not establish how many tools each page contains, so it cannot be converted into a maximum number of tools. Nor does this SDK behavior establish the limit for other MCP clients. If a catalog appears incomplete, check whether the client traversed all pages and whether its aggregation settings stopped the process.
Rank #2
- Built for Local AI and Advanced Workflows – The BOSGAME M5 AI Mini PC is powered by AMD Ryzen AI Max+ 395 with 16 cores, 32 threads, up to 5.1GHz, 50 TOPS NPU performance and up to 126 TOPS total AI performance. It is designed for local AI inference, private AI assistants, coding, data analysis, virtualization, content creation and demanding multitasking while keeping sensitive data on the device.
- 128GB Unified Memory for Large Models and Creative Projects – M5 includes 128GB LPDDR5X-8000 unified memory, giving the CPU and Radeon 8060S graphics access to a large shared memory pool. This helps support memory-intensive AI workloads, large project files, multiple virtual machines, 3D work, video editing and complex professional applications without the capacity limits of typical 32GB or 64GB mini computers.
- Radeon 8060S Graphics for Creation, Rendering and Gaming – Integrated Radeon 8060S graphics with 40 RDNA 3.5 compute units delivers high-end visual performance without a separate graphics card. Use the M5 creator workstation for 4K video editing, 3D rendering, CAD, AI image workflows, high-resolution media and modern gaming, while maintaining a compact desktop footprint.
- 2TB PCIe 4.0 SSD and Flexible Expansion – A pre-installed 2TB NVMe PCIe 4.0 SSD provides fast access to models, datasets, media libraries and project files. A second M.2 2280 PCIe 4.0 slot allows additional storage expansion, while the SD 4.0 card reader supports efficient photo and video workflows for creators and production teams.
- Professional Connectivity and Four-Display Support – Dual USB4 ports, HDMI 2.1 and DisplayPort 1.4 support up to four displays and resolutions up to 8K@60Hz. WiFi 7, Bluetooth 5.4 and 2.5GbE deliver fast networking for cloud collaboration, NAS access and business deployment. Windows 11 Pro, performance-mode switching, Wake-on-LAN and auto power-on support flexible workstation use.
Why more tools can make selection harder
Even when discovery succeeds, a model can struggle to choose among a growing set of candidates—especially when descriptions overlap or tools have similar names. One 2026 paper, MCP Server Architecture Patterns for LLM-Integrated Applications, reports selection accuracy below 90% at 10–15 tools for Claude Haiku 4.5 and at 20–30 tools for Sonnet 4. Those are results reported for the paper’s tested settings, not a universal threshold, an MCP requirement, or evidence that every model fails at those catalog sizes. The available result does not establish that the findings generalize to other models, descriptions, tasks, or hosts. See the arXiv paper listing.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesHow to diagnose a 90-plus-tool problem
- Confirm the failure mode. Determine whether the server errors, the client returns an incomplete list, the host omits tools from the model’s available choices, or the model selects the wrong tool. Record the exact host and SDK versions and a reproducible task.
- Check page traversal. Inspect the client’s
tools/listrequests and cursors. Confirm that it requests subsequent pages and does not stop at an implementation-specific page limit. If using the TypeScript SDK’s automatic aggregation, account for its documented default of 64 pages. - Compare the server’s available tools with the client’s catalog. Check authorization context and the returned pages. The specification permits availability to vary according to authorization, so compare like with like.
- Test selection separately from discovery. If every tool is listed but the wrong one is chosen, try representative tasks with the actual tool descriptions and task mix. A successful list response does not prove reliable model selection.
- Reduce the candidate set and retest. Apply filtering or on-demand discovery, then compare whether the same tasks become easier to route. Treat this as an implementation test, not proof of a protocol limit.
Ways to manage a large catalog
Filter before registering tools
Application-side filtering can register only the tools relevant to a user, workflow, or task. This reduces the number of candidates presented at once, but requires the application to choose and maintain the filtering rules. AWS guidance describes filtering as one approach to managing large tool sets: AWS machine-learning guidance.
Rank #3
- FAST RUNS IN THE FAMILY — The 14-inch MacBook Pro with the M5 Pro or M5 Max chip brings next-generation speed and powerful on-device AI to personal, professional, and creative tasks. With all-day battery life, double the starting storage,* and a breathtaking Liquid Retina XDR display, it’s pro in every way.*
- BUCKLE UP — Along with a next-generation CPU, faster unified memory, and up to 2x faster SSD storage,* M5 Pro and M5 Max feature a more powerful GPU with a Neural Accelerator built into each core, delivering faster AI performance and on-device training capabilities. So you can blaze through demanding workloads at mind-bending speeds.
- BUILT FOR AI — Apple silicon, and every major component that powers it, is designed to run demanding on-device AI workloads like LLM inference and training. And Apple Intelligence helps you write, express yourself, and get things done effortlessly with groundbreaking privacy protections at every step.*
- ALL-DAY BATTERY LIFE — MacBook Pro delivers the same exceptional performance whether it’s running on battery or plugged in.*
- MACOS RUNS APPS FAST — All your go-to apps run lightning fast in macOS, including built-in apps like FaceTime and Messages. Plus, built-in virus protection and free software updates help keep your Mac running smoothly and securely.
Retrieve tools on demand
Instead of exposing the full catalog for every request, an application can first identify a relevant subset and make those tools available when needed. This can reduce unnecessary catalog exposure, but introduces a discovery step that must itself retrieve an appropriate set.
Use semantic search to shortlist candidates
A semantic-search layer can match a task to likely tools and present a smaller candidate set. AWS guidance also describes this pattern. Its quality depends on the descriptions, retrieval method, and task; it should be evaluated against real requests rather than assumed to work for every catalog.
Rank #4
- 💥【AI 9 HX 470 GAMING PC】The BOSGAME VTA-439 mini pc is powered by AMD Ryzen AI 9 HX 470 (12C/24T, 5.2GHz) with XDNA 2 NPU: 55 TOPS dedicated AI, 86 TOPS total platform performance. Run local LLMs, AI image generation, 8K video, and 3D rendering with zero cloud latency and full privacy. Copilot+ PC certified – the ultimate AI workstation for developers and creators.
- 💥【32GB to 256GB RAM + 1TB to 8TB SSD】The BOSGAME ai mini gaming pc comes with 32GB DDR5 5600MHz RAM (dual slots max 256GB) and 1TB PCIe 4.0 SSD (triple M.2 NVMe slots max 8TB total). Each RAM max 64GB; each SSD slot max 4TB. -Upgrade anytime as your needs grow, multitask working can be performed smoothly.
- 💥【OCULINK eGPU PORT】The Oculink port provides a dedicated PCIe 4.0 x4 connection with up to 64 Gbps bandwidth—significantly higher than Thunderbolt 4's 32 Gbps PCIe data bandwidth. This direct connection delivers better frame rates and lower latency for external GPU setups, giving gamers and content creators the performance edge they need.
- 💥【DUAL 2.5GbE + Wi-Fi 7 + BT 5.4】Dual 2.5GbE LAN ports enable firewall, link aggregation, soft routing, and NAS applications. Built-in Wi-Fi 7 and Bluetooth 5.4 offer stable, high-speed wireless connections for projectors, printers, monitors, speakers, and more—ideal for a versatile, clutter-free workspace.
- 💥【RADEON 890M GPU & QUAD-SCREEN DISPLAY】Integrated with AMD Radeon 890M graphics running at 3100 MHz, the ai pc supports quad display output via HDMI 2.1 (4K@144Hz), DP 1.4 (4K@144Hz), USB4 (8K@60Hz), and Full-Function Type-C. Perfect for AAA gaming, video editing, 3D modeling, and multitasking—deliver stunning visuals across four screens with fluid performance.
What to measure before calling it a 90-tool limit
- Protocol pagination: Does the server return pages, and does the client follow the cursors?
- Client behavior: Does the host or SDK impose an aggregation cap or cache results? What happens when the catalog changes?
- Catalog exposure: Are all tools registered for every request, or are they filtered or retrieved when needed?
- Selection accuracy: For the actual descriptions and tasks, does the model choose the intended tool?
- Reproducibility: Can the issue be tied to a specific server, client, SDK version, authorization context, and number of pages or tools?
Without those details, “the catalog breaks at 90” is too vague to diagnose and should not be treated as a claim about MCP itself.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
