Skip to main content
Glama

osirAgentDeploy

osirAgentDeploy: Install a ready-made AI agent - Hermes (Hermes Agent by Nous Research, a self-hosted personal AI agent with a web dashboard) - on the user's OWN VPS, behind HTTPS with a password-protected dashboard. Use this for 'run/install Hermes on my server' INSTEAD of osirAppDeploy, a hand-written Dockerfile, SSH or an install script: the platform installs the vetted agent itself. Needs a VPS with at least 4 GB RAM. TWO WAYS IN, like osirAppDeployToVps: (1) the user already owns a VPS: pass instanceId (from listMyVpsInstances) - spends nothing, no confirmation; (2) no server yet: pass packageId (from listVpsPackages, 4 GB+ RAM) - stages a VPS order (COSTS MONEY): returns an actionId; present the price/summary and call executeConfirmedAction only if the user approves. The result carries the username and the PASSWORD: show the password to the user once, in that reply - it is never shown again (osirAgentResetPassword makes a new one). Then poll osirAgentStatus until installed - it gives the dashboard URL (without a domain, Osir assigns the address; never build it) - and connect a model provider with osirAgentConfigure. Calling this again resumes an install and never buys a second VPS. If the platform says the agent is not available yet, tell the user plainly. 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": {
      "agent": "<agent>"
    },
    "name": "osirAgentDeploy"
  }
}

Input Schema

NameRequiredDescriptionDefault
agentYesThe agent to install. Currently only 'hermes'.
domainNoA domain the user owns to serve the dashboard on, e.g. 'agent.example.com'. Bound automatically if its DNS is hosted at Osir, otherwise the result says which A record to add. Omit it to get an Osir address automatically; never pass an osir.app name.
usernameNoDashboard login name; default 'admin'.
packageIdNoVPS package id from listVpsPackages (at least 4 GB RAM). Only when a server has to be ordered; omit it when passing instanceId.
instanceIdNoId of a VPS the user ALREADY owns, from listMyVpsInstances. Nothing is ordered or charged. Never invent one.
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.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / domain / description
      Previous value: -"Domain to serve the dashboard on, e.g. 'agent.example.com'. Bound automatically if its DNS is hosted at Osir, otherwise the result says which A record to add."New value: +"A domain the user owns to serve the dashboard on, e.g. 'agent.example.com'. Bound automatically if its DNS is hosted at Osir, otherwise the result says which A record to add. Omit it to get an Osir address automatically; never pass an osir.app name."
  2. Added

TDQS

A4.9/5.0
Behavior5/5

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

Far beyond the annotations: discloses that the packageId path spends money and stages an order requiring confirmation, that the result contains a password shown only once, that repeat calls resume an install rather than buying a second VPS, and that authentication is required. The resume-on-repeat behavior is a mild tension with idempotentHint=false but not a contradiction — it describes progress continuation, not identical repeated effects.

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?

Front-loaded with purpose and the primary routing decision, and nearly every clause carries operational weight (cost, secret handling, polling). It is dense and parenthetical-heavy — the Hermes vendor blurb and some inline asides could be trimmed — but no sentence is filler.

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?

With no output schema, the description carries the return-value burden itself: actionId from the staged order, username and one-time password, dashboard URL only via osirAgentStatus (never constructed manually), and the failure mode where the platform reports the agent unavailable. Nothing needed to invoke or follow up correctly is missing.

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

Parameters5/5

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

Schema coverage is 100%, yet the description still adds decision-relevant meaning: instanceId 'spends nothing, no confirmation' versus packageId 'COSTS MONEY' returning an actionId, the 4 GB RAM constraint on packages, and the domain auto-binding/Osir-assigned-address rule. This is cost and side-effect semantics the schema cannot express.

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 (install the Hermes AI agent on the user's own VPS behind HTTPS) and immediately names the alternatives it is not (osirAppDeploy, hand-written Dockerfile, SSH, install script). An agent can distinguish this from the sibling app-deploy tools without opening any 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?

Explicitly splits usage into two mutually exclusive paths — existing VPS via instanceId (no cost, no confirmation) vs. no server via packageId (costs money, must call executeConfirmedAction after approval) — and states the prerequisite (4 GB+ RAM) plus the downstream sequence (osirAgentStatus polling, then osirAgentConfigure). It also gives an exclusion: never pass an osir.app name as domain.

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.