Skip to main content
Glama

osirAppDeploy

osirAppDeploy: Deploy an app to Osir (free tier) and get a live HTTPS URL; the app runs isolated in a microVM. Send the project's source as 'files' (each {path, content}) - you never zip, upload or make an HTTP request yourself. Returns immediately with status BUILDING: poll osirAppStatus until READY (typically 1-2 min, up to ~7 for a first build) before telling the user the URL is live; if it ends BUILD_FAILED, fix the source and redeploy - never deploy around a failed build by hand. Deploying an existing app name redeploys it (new version) and applies any secrets set via osirAppSetSecret. A plain static website (HTML/CSS/JS with no framework or build step) is also supported: it is auto-detected and served directly; pass language 'node' for it. If the project has its own Dockerfile at the root, the platform builds exactly that: pass language 'node' for it; the container must listen on port 8080. Shared-tier apps get 256 MB / 0.5 CPU, so a big image-based app belongs on the user's own VPS (osirAppDeployToVps). For a ready-made AI agent (Hermes), use osirAgentDeploy - never this tool and never a hand-written Dockerfile. To put the app on the user's OWN VPS, server or production box instead, deploy it here first and then call osirAppDeployToVps - never SSH into the box or write an install script. Once an app is on the user's own VPS, redeploying HERE under the same name is how you update it: the platform re-ships it to the box and keeps its domain. Requires authentication.

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>",
      "language": "<language>"
    },
    "name": "osirAppDeploy"
  }
}

Input Schema

NameRequiredDescriptionDefault
nameYesApp name: lowercase letters, digits, and hyphens, e.g. 'habit-tracker'.
filesNoThe project's source files: [{path, content, encoding?}]. path is relative and forward-slashed (e.g. 'src/index.js'); encoding is 'utf8' (default) or 'base64' for binaries such as images. Include every file the app needs - omitted files are not in the new version. Up to 500 files / 6 MiB. Use this OR uploadTicket.
regionNoRegion: 'us' or 'al' ('al' is Albania/Tirana); defaults to the platform's home region.
languageYesRuntime language: 'node', 'python', 'php-laravel', or 'go'; use 'node' for a plain static site or a project with its own Dockerfile.
sessionKeyNoSession key (osk_...) returned by checkDeviceLoginStatus. Pass it on every call when logged in via the in-chat device flow; omit when connected via OAuth.
uploadTicketNoOnly for the shell path: upload ticket from osirAppCreateUpload, after PUTting the zipped source to its putUrl. Leave out when sending files.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changed
    • addedInput schema / properties / files
      Added value: +{
      +  "description": "The project's source files: [{path, content, encoding?}]. path is relative and forward-slashed (e.g. 'src/index.js'); encoding is 'utf8' (default) or 'base64' for binaries such as images. Include every file the app needs - omitted files are not in the new version. Up to 500 files / 6 MiB. Use this OR uploadTicket.",
      +  "items": {
      +    "properties": {
      +      "content": {
      +        "type": "string"
      +      },
      +      "encoding": {
      +        "type": "string"
      +      },
      +      "path": {
      +        "type": "string"
      +      }
      +    },
      +    "type": "object"
      +  },
      +  "type": "array"
      +}
    • changedInput schema / properties / language / description
      Previous value: -"Runtime language: 'node', 'python', 'php-laravel', or 'go'; use 'node' for a plain static site."New value: +"Runtime language: 'node', 'python', 'php-laravel', or 'go'; use 'node' for a plain static site or a project with its own Dockerfile."
    • changedInput schema / properties / uploadTicket / description
      Previous value: -"Upload ticket from osirAppCreateUpload, after uploading the zipped source to its putUrl."New value: +"Only for the shell path: upload ticket from osirAppCreateUpload, after PUTting the zipped source to its putUrl. Leave out when sending files."
    • changedInput schema / required
      Previous value: -[
      -  "name",
      -  "language",
      -  "uploadTicket"
      -]New value: +[
      +  "name",
      +  "language"
      +]
  2. Changed4 schema fields changed
    • addedInput schema / properties / language / description
      Added value: +"Runtime language: 'node', 'python', 'php-laravel', or 'go'; use 'node' for a plain static site."
    • addedInput schema / properties / name / description
      Added value: +"App name: lowercase letters, digits, and hyphens, e.g. 'habit-tracker'."
    • addedInput schema / properties / region / description
      Added value: +"Region: 'us' or 'al' ('al' is Albania/Tirana); defaults to the platform's home region."
    • addedInput schema / properties / uploadTicket / description
      Added value: +"Upload ticket from osirAppCreateUpload, after uploading the zipped source to its putUrl."
  3. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Beyond the annotations (readOnlyHint=false, idempotentHint=false), it discloses that the call returns immediately with status BUILDING, typical and worst-case build times (~1-2 min, up to ~7 for a first build), the BUILD_FAILED outcome and its remedy, redeploy semantics for an existing name, that secrets from osirAppSetSecret are applied, and the 256 MB / 0.5 CPU shared-tier limits. It also states authentication is required and that the agent never zips, uploads, or makes HTTP requests itself.

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?

It is front-loaded with purpose, then outcome, then routing, then constraints, and nearly every sentence carries operational information. It loses a point for repetition: the instruction to pass language 'node' appears twice for two different cases, and the 'never X' prohibitions stack up, making the middle dense to parse.

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?

For a non-idempotent, asynchronous deploy tool with no output schema, it covers the async lifecycle, failure handling, redeploy/update semantics, resource limits, auth, and sibling routing. An agent has everything needed to call it, poll it, and recover from failure without further clarification.

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 already 100%, so the baseline is 3, but the description adds real meaning: 'files' carries the whole project source with no zip or upload step, 'language' should be 'node' both for a plain static site and for a project with its own root Dockerfile, and the container must listen on port 8080. The only gap is that region, sessionKey, and uploadTicket are left entirely to the schema.

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?

The description opens with a specific verb and resource ('Deploy an app to Osir') plus the outcome ('get a live HTTPS URL') and the isolation model ('runs isolated in a microVM'). It explicitly differentiates from siblings by naming osirAppDeployToVps and osirAgentDeploy and stating which cases belong to them.

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?

It gives explicit routing rules: use osirAgentDeploy for a ready-made Hermes agent, use this first then osirAppDeployToVps for the user's own VPS, and redeploy here under the same name to update a VPS-hosted app. It also states the post-call procedure (poll osirAppStatus until READY) and the failure path (fix source and redeploy rather than hand-deploying around a failed build).

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.