Skip to main content
Glama
meshive-mcp

meshive-mcp

Official
by meshive

Estimate Pod Price

estimate_pod
Read-onlyIdempotent

Preview a pod's hourly price and hardware before creating it, with no pod created and no credit spent. Check the cost estimate, then create the pod with the same arguments.

Instructions

Estimate the hourly price of a pod before creating it; nothing is created and no credit is spent. Call this first, show the user the price and hardware, then call create_pod with the same arguments. locked_ports and open_ports list the ports the pod would expose over HTTPS, by the lock each would get (open_by_template: open because the template sets them so, not because of this call).

Each argument's format is in its own description. workspace and template_id come from the workspaces and templates tools. GPU pod: pass gpu_model (from the gpus tool) and optionally gpu_count/gpu_vram_gb; vCPU/RAM default to the GPU's recommended sizes. CPU pod: omit gpu_model; cpu_line picks the CPU line (from the cpus tool; default the cheapest line that fits) and vCPU/RAM default to 8 cores / 32 GiB like the console (or the nearest sizes that line offers). vcpu and ram_gb are totals for the pod and must be sizes the console offers. The system disk is sized by the server (shown as resources.disk_gb in the estimate). volumes attaches existing network storage. local_volumes creates new local storage on the pod's machine, billed hourly by size until it is deleted, also while the pod is stopped and after the pod is deleted unless delete_pod lists it. max_price_per_hour caps the compute hourly rate; existing storage you attach and Asset Hub retention are billed separately. input_assets places Asset Hub assets in the pod. New files in the template's output folders and in watched_folders are uploaded as assets; watched_template_folders changes the template's folders and harvest_destination where the files go. model_set adds one model set to the pod, and hf_token_id is the token a set gated on Hugging Face needs; the set's files are added before the pod is created, often several minutes. Ports exposed over HTTPS are created locked: only workspace members signed in to the console and requests with one of the port's access keys get through, and anything else gets 401. lock_public_ports=false (or "locked": false on a port) creates them open to anyone who knows the URL: only when the user explicitly asks for that. One exception: a template port whose template sets it open is created open; the preview lists it as open_by_template and says how to lock it.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "name": "<name>",
      "template_id": "<template_id>"
    },
    "name": "estimate_pod"
  }
}

Input Schema

NameRequiredDescriptionDefault
envNoEnvironment variables; template variables the template leaves blank need a value, except a Hugging Face token (HF_TOKEN): it's only needed for gated or private repos or Hub uploads and is left out when blank, so don't ask for one to run a public model. `secret_keys` names the secret ones.
nameYes
vcpuNovCPUs for the whole pod, one of the sizes the console offers.
portsNoOpens new ports: a list like [8888, {"port": 6006, "name": "tb", "external": false}]. A port is exposed over HTTPS unless "external" is false, and an exposed port is locked unless its item says "locked": false (open to anyone who knows the URL; only when the user explicitly asks). Don't repeat ports or port names the template already has.
ram_gbNoRAM for the whole pod in GiB, one of the sizes the console offers.
regionNoAccepted for compatibility and ignored.
commandNoOverrides the start command (at most 32,000 characters).
volumesNoAttaches EXISTING network (nfs) storage from the storages tool: [{"storage": pv_name, "mount_path": "/data"}]. A display name works on estimates/previews; confirm with the returned ids. Existing local storage can't be attached to a new pod.
cpu_lineNoCPU pods only: a `cpu_line` value from the cpus tool. Default: the cheapest line that fits.
gpu_countNo
gpu_modelNoGPU model from the gpus tool. Omit it for a CPU pod.
model_setNoAdds one model set to the pod: a `slug` from the model_sets tool for this template (only templates whose list is `selectable`; one set per pod). Its files are added as input assets before the pod is created, so creating takes longer, often several minutes. The estimate shows the set as `model_set` with its file count, size and minimum GPU memory. If a file can't be added, no pod is created (model_set_unavailable) and the files already added are reused next time.
workspaceNoWorkspace id from the workspaces tool (a label also works).
cpu_premiumNo
gpu_vram_gbNo
hf_token_idNoFor Quick Deploy templates whose model set is gated on Hugging Face (the model_sets tool shows `hf_token_required` and `gated_sources`): the id of a workspace token from the source_credentials tool whose Hugging Face account accepted the model's license. The set's files download with that token, so the user must first accept the license on the page in `gated_sources` with the same Hugging Face account. Like `model_set`, a confirmed create with it returns only after the files were linked, which can take minutes.
rental_typeNodemand
secret_keysNoNames of the `env` variables that are secret.
template_idYesTemplate id from the templates tool.
input_assetsNoPlaces Asset Hub assets in the pod: ["asset_id" or {"asset": id, "target_dir": "/workspace/models", "role": role, "paths": ["sub/folder/"], "strip_prefix": ""}] (default place /inputs/<asset name> or the template's path for that role); "paths" picks some files or folders, copied from the asset's file list exactly as written there, and "strip_prefix": "" keeps their folder structure (left out, the folder they share is stripped). Only these keys are accepted. They add an input volume billed by size.
local_volumesNoCreates NEW local storage on the pod's machine together with the pod: [{"name": "scratch", "mount_path": "/scratch", "size_gb": 50}] with optional "encrypted", "usage_warning" and "usage_warning_threshold" (0-1). Each needs at least 10 GiB and spot pods can't have one. It is billed hourly by size until it is deleted, also while the pod is stopped and after the pod is deleted unless delete_pod lists it; the estimate includes it as `breakdown.local_storage`.
template_portsNoChanges the ports the template opens, like the console's deploy screen: [{"port": 8188, "name": "ui", "new_port": 8189, "external": false}], where `port` is the template's port and "name" (rename), "new_port" (renumber), "external" (true = exposed over HTTPS, false = internal only) and "locked" (for an exposed port: true = locked, also when the template sets that port open; false = open to anyone who knows the URL, only when the user explicitly asks) are optional. Template ports can't be removed; use `ports` to open new ones.
uptime_premiumNo
watched_foldersNoUploads new files in these folders as assets: ["/workspace/results" or {"path": p, "include": ["*.png"], "include_existing": false, "enabled": true}] (the template's own output folders are watched already). It works on templates without an output folder too (for example NIM or vLLM), as long as the pod has storage the folders sit in (`volumes` or `local_volumes`).
internet_premiumNo
lock_public_portsNoLeave it true (the default): the ports the pod exposes over HTTPS are created locked, so only workspace members signed in to the console and requests that carry one of the port's access keys get through. The one exception is a template port that its template sets open: it stays open unless its `template_ports` item says "locked": true (the preview lists such ports as `open_by_template`). false creates every exposed port open to anyone who knows the URL: pass it only when the user explicitly asked for open ports. "locked" on a port's item in `ports` or `template_ports` decides for that port alone.
max_price_per_hourNoChecked against the estimate (compute plus new local_volumes) before creating, and caps the final COMPUTE hourly rate at placement; an over-cap placement fails asynchronously. Existing storage you attach and Asset Hub retention are billed separately and excluded from this cap.
harvest_destinationNo{"mode": "user_s3", "credential_id": n} sends the uploaded files to the user's bucket instead of managed storage.
watched_template_foldersNoChanges the template's output folders from the start, like the console's create screen: {"/outputs": {"enabled": false}} turns one off, {"/outputs": {"include": ["*.png"]}} uploads only matching files, and {"/workspace/output": {"path": "/workspace/results"}} moves an output folder (its $MESHIVE_OUTPUT_DIR and storage follow; output folders only).

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed22 schema fields changedv1.10.0
    • addedInput schema / properties / command / description
      Added value: +"Overrides the start command (at most 32,000 characters)."
    • addedInput schema / properties / cpu_line / description
      Added value: +"CPU pods only: a `cpu_line` value from the cpus tool. Default: the cheapest line that fits."
    • addedInput schema / properties / env / description
      Added value: +"Environment variables; template variables the template leaves blank need a value, except a Hugging Face token (HF_TOKEN): it's only needed for gated or private repos or Hub uploads and is left out when blank, so don't ask for one to run a public model. `secret_keys` names the secret ones."
    • addedInput schema / properties / gpu_model / description
      Added value: +"GPU model from the gpus tool. Omit it for a CPU pod."
    • addedInput schema / properties / harvest_destination / description
      Added value: +"{\"mode\": \"user_s3\", \"credential_id\": n} sends the uploaded files to the user's bucket instead of managed storage."
    • addedInput schema / properties / hf_token_id / description
      Added value: +"For Quick Deploy templates whose model set is gated on Hugging Face (the model_sets tool shows `hf_token_required` and `gated_sources`): the id of a workspace token from the source_credentials tool whose Hugging Face account accepted the model's license. The set's files download with that token, so the user must first accept the license on the page in `gated_sources` with the same Hugging Face account. Like `model_set`, a confirmed create with it returns only after the files were linked, which can take minutes."
    • addedInput schema / properties / input_assets / description
      Added value: +"Places Asset Hub assets in the pod: [\"asset_id\" or {\"asset\": id, \"target_dir\": \"/workspace/models\", \"role\": role, \"paths\": [\"sub/folder/\"], \"strip_prefix\": \"\"}] (default place /inputs/<asset name> or the template's path for that role); \"paths\" picks some files or folders, copied from the asset's file list exactly as written there, and \"strip_prefix\": \"\" keeps their folder structure (left out, the folder they share is stripped). Only these keys are accepted. They add an input volume billed by size."
    • addedInput schema / properties / local_volumes / description
      Added value: +"Creates NEW local storage on the pod's machine together with the pod: [{\"name\": \"scratch\", \"mount_path\": \"/scratch\", \"size_gb\": 50}] with optional \"encrypted\", \"usage_warning\" and \"usage_warning_threshold\" (0-1). Each needs at least 10 GiB and spot pods can't have one. It is billed hourly by size until it is deleted, also while the pod is stopped and after the pod is deleted unless delete_pod lists it; the estimate includes it as `breakdown.local_storage`."
    • addedInput schema / properties / lock_public_ports
      Added value: +{
      +  "default": true,
      +  "description": "Leave it true (the default): the ports the pod exposes over HTTPS are created locked, so only workspace members signed in to the console and requests that carry one of the port's access keys get through. The one exception is a template port that its template sets open: it stays open unless its `template_ports` item says \"locked\": true (the preview lists such ports as `open_by_template`). false creates every exposed port open to anyone who knows the URL: pass it only when the user explicitly asked for open ports. \"locked\" on a port's item in `ports` or `template_ports` decides for that port alone.",
      +  "title": "Lock Public Ports",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / max_price_per_hour / description
      Added value: +"Checked against the estimate (compute plus new local_volumes) before creating, and caps the final COMPUTE hourly rate at placement; an over-cap placement fails asynchronously. Existing storage you attach and Asset Hub retention are billed separately and excluded from this cap."
    • addedInput schema / properties / model_set / description
      Added value: +"Adds one model set to the pod: a `slug` from the model_sets tool for this template (only templates whose list is `selectable`; one set per pod). Its files are added as input assets before the pod is created, so creating takes longer, often several minutes. The estimate shows the set as `model_set` with its file count, size and minimum GPU memory. If a file can't be added, no pod is created (model_set_unavailable) and the files already added are reused next time."
    • addedInput schema / properties / ports / description
      Added value: +"Opens new ports: a list like [8888, {\"port\": 6006, \"name\": \"tb\", \"external\": false}]. A port is exposed over HTTPS unless \"external\" is false, and an exposed port is locked unless its item says \"locked\": false (open to anyone who knows the URL; only when the user explicitly asks). Don't repeat ports or port names the template already has."
    • addedInput schema / properties / ram_gb / description
      Added value: +"RAM for the whole pod in GiB, one of the sizes the console offers."
    • addedInput schema / properties / region / description
      Added value: +"Accepted for compatibility and ignored."
    • addedInput schema / properties / secret_keys / description
      Added value: +"Names of the `env` variables that are secret."
    • addedInput schema / properties / template_id / description
      Added value: +"Template id from the templates tool."
    • addedInput schema / properties / template_ports / description
      Added value: +"Changes the ports the template opens, like the console's deploy screen: [{\"port\": 8188, \"name\": \"ui\", \"new_port\": 8189, \"external\": false}], where `port` is the template's port and \"name\" (rename), \"new_port\" (renumber), \"external\" (true = exposed over HTTPS, false = internal only) and \"locked\" (for an exposed port: true = locked, also when the template sets that port open; false = open to anyone who knows the URL, only when the user explicitly asks) are optional. Template ports can't be removed; use `ports` to open new ones."
    • addedInput schema / properties / vcpu / description
      Added value: +"vCPUs for the whole pod, one of the sizes the console offers."
    • addedInput schema / properties / volumes / description
      Added value: +"Attaches EXISTING network (nfs) storage from the storages tool: [{\"storage\": pv_name, \"mount_path\": \"/data\"}]. A display name works on estimates/previews; confirm with the returned ids. Existing local storage can't be attached to a new pod."
    • addedInput schema / properties / watched_folders / description
      Added value: +"Uploads new files in these folders as assets: [\"/workspace/results\" or {\"path\": p, \"include\": [\"*.png\"], \"include_existing\": false, \"enabled\": true}] (the template's own output folders are watched already). It works on templates without an output folder too (for example NIM or vLLM), as long as the pod has storage the folders sit in (`volumes` or `local_volumes`)."
    • addedInput schema / properties / watched_template_folders / description
      Added value: +"Changes the template's output folders from the start, like the console's create screen: {\"/outputs\": {\"enabled\": false}} turns one off, {\"/outputs\": {\"include\": [\"*.png\"]}} uploads only matching files, and {\"/workspace/output\": {\"path\": \"/workspace/results\"}} moves an output folder (its $MESHIVE_OUTPUT_DIR and storage follow; output folders only)."
    • addedInput schema / properties / workspace / description
      Added value: +"Workspace id from the workspaces tool (a label also works)."
  2. First observedv0.1.0

TDQS

A4.7/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint/idempotent/destructive=false, but the description adds behavior the annotations cannot: no credit is spent, ports are created locked with 401 for unauthenticated requests, local_volumes keep billing hourly while stopped and after pod deletion, model_set files are added before creation and can take minutes, and max_price_per_hour caps only compute while attached storage and Asset Hub retention bill separately. These are exactly the side-effect and cost facts an agent needs before calling.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The opening two sentences are excellent and front-loaded, but the body is long and substantially restates schema parameter descriptions nearly verbatim (lock_public_ports, local_volumes, model_set, hf_token_id, input_assets, template_ports, max_price_per_hour). That duplication inflates length without adding meaning beyond the structured fields, though the paragraph-per-topic structure keeps it navigable.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values need not be explained, and the description still points at the relevant estimate fields (resources.disk_gb, breakdown.local_storage, model_set file count/size, open_by_template). For a 29-parameter read-only estimator, the workflow, defaults, billing caveats, and port-security model are all covered.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 29 parameters at 76% schema coverage, the description meaningfully compensates: it explains the GPU-vs-CPU argument split, that vcpu/ram_gb are pod totals in console-offered sizes, that volumes attach existing NFS while local_volumes create new storage, and the locked/open port semantics. A few parameters (rental_type, cpu_premium, uptime_premium, internet_premium) are undocumented in both schema and description, so it stops short of full coverage.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ("Estimate the hourly price of a pod") plus the crucial scope qualifier that nothing is created and no credit is spent. It explicitly positions itself against the sibling create_pod ("Call this first ... then call create_pod with the same arguments"), so an agent can distinguish the two without opening either schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives an explicit ordered workflow: estimate first, show the user price and hardware, then create with the same arguments. It also branches on pod type (GPU pod: pass gpu_model; CPU pod: omit it and pick cpu_line) and states the default-selection rule (cheapest line that fits). This is when-to-use guidance plus alternatives, not just context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.