Log inSubmit server
Kosha Model Discovery logo

Kosha Model Discovery

Discovers models, pricing, credentials, routes, and capabilities across providers and local runtimes through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Kosha Model Discovery

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "kosha-model-discovery": {
      "command": "npx",
      "args": [
        "-y",
        "-p",
        "@sriinnu/kosha-discovery",
        "kosha-mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

Kosha Model Discovery MCP server exposes model discovery and routing data to agents over stdio. Its tools query models across 46 providers and local runtimes, resolve aliases, compare prices, inspect serving routes, check provider health, and select context strategies. It can use available credentials for richer discovery, while public models.dev and LiteLLM catalogs plus an offline list support operation without API keys. Reach for it when an agent needs to choose a model based on cost, capabilities, context limits, or available credentials.

Use cases

  • Find the cheapest model that supports a required capability and context window
  • Resolve model aliases to canonical provider IDs
  • Compare serving routes for a selected model
  • Check provider health before choosing a route

Key features

  • Model discovery across 46 providers and local runtimes
  • Pricing, capability, and context-limit lookup
  • Credential-aware model selection
  • Alias resolution
  • Route ranking and route inspection
  • Provider health and context strategy tools

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

7/7 checks healthy over the last 26d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 9d ago via OSV.dev · @sriinnu/kosha-discovery (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars3Stargazers on the source repository.
npm downloads3.1kPackage downloads in the last 30 days.
Last commit1d agoMost recent push to the default branch.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Kosha Model Discovery works

What Kosha Model Discovery does

Kosha Model Discovery gives MCP-compatible agents access to a registry of AI models, providers, pricing, capabilities, context limits, credentials, and serving routes. It is designed for decisions such as finding the least expensive model that supports tool use and a required context window, or determining which route is available for a model alias.

The package also includes a TypeScript library, CLI, HTTP API, OpenAI-compatible proxy, and spend ledger, but the MCP entry point is kosha-mcp. The MCP server runs over standard input and output, so an agent can use it without exposing a separate HTTP service.

How it works

On startup and during discovery, Kosha checks configured provider access and local credential sources, then combines the results with public catalogs from models.dev and LiteLLM and a curated offline model list. Provider keys are optional. With no credentials, the server can still answer from publicly available model metadata; available credentials allow it to identify models the user can access.

The registry covers 46 providers and local runtimes. Model records include pricing and context information where available, along with capabilities and provider relationships. Alias resolution lets an agent use names such as sonnet and obtain the corresponding canonical model ID.

Setup and configuration

Install the package with Node.js 22 or newer:

Terminal
npm install @sriinnu/kosha-discovery

Register the stdio server with Claude using:

Terminal
claude mcp add kosha -- kosha-mcp

Clients that read the MCP registry can install the published server identifier io.github.sriinnu/kosha-discovery. Provider credentials are optional, and the server can run on a fresh machine using its public catalogs and offline data.

Tools and capabilities

The MCP tool set includes:

  • kosha_query_models for searching the model registry.
  • kosha_cheapest_model for finding lower-cost models that meet requirements.
  • kosha_ranked_routes for ranking available serving paths.
  • kosha_model_detail for inspecting a model record.
  • kosha_model_routes for listing routes for a model.
  • kosha_resolve_alias for mapping an alias to a canonical model ID.
  • kosha_provider_health for checking provider status.
  • kosha_context_strategy for selecting a model or approach based on context needs.

These tools expose the discovery portion of Kosha to an agent. The separate proxy can resolve aliases, choose providers, forward OpenAI-compatible requests, and record spend, but those HTTP proxy operations are not required for the MCP server.

Limitations and notes

Discovery quality depends on the catalogs and credentials available to Kosha. Google, Bedrock, and Vertex are listed as discovery-only in the broader package, while some other non-chat provider formats are also not proxied. The MCP server itself uses stdio, so clients must support launching a local MCP process. Provider credentials are not required for basic public-catalog queries, but credential-aware selection depends on the relevant credentials being available to Kosha.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Kosha Model Discovery

Kosha Model Discovery is an MCP server that connects agents to model catalogs across 46 providers and local runtimes. Its main tools query models, find cheaper options, rank routes, inspect model details, resolve aliases, check provider health, and choose context strategies. It runs over stdio and works without provider keys using public catalogs.

Install it with `npm install @sriinnu/kosha-discovery`, then register it with `claude mcp add kosha -- kosha-mcp`. Node.js 22 or newer is required. Clients that consume the MCP registry can use the published identifier `io.github.sriinnu/kosha-discovery`.

No, API keys are not required for basic discovery. Without credentials, Kosha uses public models.dev and LiteLLM catalogs together with a curated offline list. Provider credentials enable credential-aware discovery and selection for models the user can access.

Use `kosha_cheapest_model` with the required role, capability, and context constraints. Kosha ranks models by price and includes credential-related information, allowing an agent to distinguish an inexpensive model from one it can actually use.

Call `kosha_resolve_alias` with the alias, such as `sonnet`, to obtain its canonical model ID. The related model-detail and route tools can then provide pricing, capabilities, provider information, and available serving paths.

Related MCP servers

All in Conversational AI All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewKosha Model Discovery AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/kosha-model-discovery?style=directory)](https://allmcps.com/mcp/kosha-model-discovery)
HTML Embed
<a href="https://allmcps.com/mcp/kosha-model-discovery"><img src="https://allmcps.com/api/badge/kosha-model-discovery?style=directory" alt="Kosha Model Discovery on AllMCPs" /></a>