estimate_self_hostingVRAM, GPUs needed, single-stream and batched throughput, hardware cost and the vendor's official API price/speed for self-hosting an open-weights model. Omit model to list available models.
| Name | Required | Description | Default |
|---|---|---|---|
| gpus | No | gpu ids from list_gpu_prices, e.g. h100_sxm, h20, ascend910b, rtx_pro_6000, rtx5090, rtx5090d | |
| model | No |
Changes observed during successful MCP inspections.
Input schema / properties / gpus / descriptionPrevious value: -"gpu ids, e.g. h100_sxm, h20, l20, ascend910b, rtx5090d"New value: +"gpu ids from list_gpu_prices, e.g. h100_sxm, h20, ascend910b, rtx_pro_6000, rtx5090, rtx5090d"Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint/true, idempotentHint/true, destructiveHint/false, and openWorldHint/false, so the safety profile is fully covered before the description speaks. The description adds the non-obvious dual behavior of the model parameter (omitting it switches the tool into a listing mode), which is genuinely useful context, but says nothing about units, estimation assumptions, or cost basis.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences with no filler, and the listing-mode instruction sits at the end where it is easy to miss but does not bloat the front. The first sentence is an output enumeration rather than a verb, so it is efficient but slightly less front-loaded on purpose.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description usefully enumerates the returned metrics (VRAM, GPU count, throughput, hardware cost, API price/speed), which is exactly what the agent needs to judge relevance. Missing are units, estimation assumptions, and any hint of result format, but for a read-only 2-parameter estimator this is close to complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Coverage is 50%: the schema documents gpus (including example ids) but leaves model bare. The description compensates by explaining model's optionality and the listing behavior when it is absent, which is meaning the schema does not carry.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Enumerates exactly what the tool computes — VRAM, GPU count, single-stream and batched throughput, hardware cost, and vendor API price/speed — for self-hosting an open-weights model, which is far more specific than restating the title. It does not explicitly distinguish itself from the sibling estimate_enterprise_ai_cost, which is the only real gap.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives one concrete usage rule ('Omit model to list available models') and implicitly routes GPU selection through list_gpu_prices, so usage is partly implied rather than stated. There is no guidance on when to choose this over estimate_enterprise_ai_cost or recommend_coding_pc.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.