Skip to main content
Glama

Krate

Build and publish a Krate app

build_app

Builds a real app the person keeps: one small file that runs natively on Windows, macOS and Linux and opens from a link in any browser, phones included. A Krate app keeps its own data on the device and works offline; with the person's permission it can print to a printer they pick with no print dialog, talk to serial and USB devices, read and write Excel files, serve other devices on the local network, and use the camera, microphone and location. Compiles the app's files (Cargo.toml, src/lib.rs, manifest.toml and any assets), checks it uses only Krate's API and runs it once, then publishes it unlisted to the signed-in person's Krate account with its link. Answers a build id at once; a build takes a few minutes. Needs the person to sign in to Krate.

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>",
      "files": "<files>"
    },
    "name": "build_app"
  }
}

Input Schema

NameRequiredDescriptionDefault
nameYesThe app's name, as a person would say it
filesYesEvery file of the app, with paths relative to the app folder

Output Schema

NameRequiredDescriptionDefault
fixNoHow to fix it, in the check's own words
linkNoThe app's own link; it opens the newest version
nameNo
stageNoWhere Krate's check stopped, on a failure
app_idNo
detailNoWhat went wrong, in the check's own words
statusYes
messageYes
build_idYes
browser_linkNoOpens the app in a browser on any device, nothing to install

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.2/5.0
Behavior4/5

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

Annotations only cover the safety flags (not read-only, not destructive, not idempotent, closed-world). The description adds genuinely non-obvious behavior: an auth requirement, an immediate id with a multi-minute build, a validation/run step before publishing, and the fact that the result is made unlisted and public-linkable on the person's account. That is real side-effect disclosure, though rate limits and failure modes are unmentioned.

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

Conciseness4/5

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

Purpose and the build id/timing note are front-loaded and the pipeline (compile → check → run → publish) is ordered clearly. The mid-section capability catalogue (print, serial, USB, Excel, LAN serving, camera/mic/location) is partially justified because it informs what the agent may write into the app, but phrases like 'the person keeps' and 'with no print dialog' are padding.

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 no explanation, yet the description still flags the async contract (id now, build in minutes). With auth, side effects, file composition and timing all covered, an agent has everything needed to call this correctly.

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?

Schema coverage is 100%, so the baseline would be 3, but the description adds meaning the schema lacks: it names which files the build expects (Cargo.toml, src/lib.rs, manifest.toml, assets), which tells the agent how to populate the generic `files` array. It adds nothing about `name` beyond the schema's own maxLength/description.

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 concrete verb and resource — compiles the app's files (Cargo.toml, src/lib.rs, manifest.toml), validates the API usage, runs it once, then publishes it unlisted to the signed-in account — and returns a build id. That separates it from build_status (which resolves an id) and get_app (which reads a published app) without either schema being opened.

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

Usage Guidelines3/5

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

Gives clear context (you have app files to build and publish) and a prerequisite ('Needs the person to sign in to Krate'), plus the async hint that a build id comes back at once while the build takes minutes. It never names the natural follow-up tool (build_status) or states when not to use this, so the routing is implied rather than explicit.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.