How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

90 days of the enterprise, pro and team plans free, no card needed. See the plans

OpenSmartRoute

eval-author - Prompt - OpenSmartRoute

Promptv1.0.0

eval-author

Name: eval-author
Author: dynamics365ninja

Draft a new eval case (eval/cases/<id>.json) + scaffold its golden folder, golden_pending until captured on the VM.

by dynamics365ninja(0) 0 installs

Free

Free account. Installing gives you the manifest plus copy-paste snippets.

See reviews

About

Prompt file imported from dynamics365ninja/d365fo-mcp-server (.claude/commands/eval-author.md). Fill in {{arguments}} before use. Copyright stays with the author.

Launch the eval-author subagent (Task tool, subagent_type: eval-author) to draft a new case for the eval catalog.

What to author: {{arguments}}

The subagent carries the authoring contract (mirror existing cases at the tier → draft via npm run eval:mine or by hand against eval/cases/schema.json → golden_pending: true → scaffold eval/goldens/<id>/ → validate schema + id/tier match + npx vitest run). It will hand off with the next step: capture the golden on the VM via the eval-implementer role. Show me the drafted case JSON.

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

Launch the **eval-author** subagent (Task tool, `subagent_type: eval-author`) to
draft a new case for the eval catalog.

What to author: **{{arguments}}**

The subagent carries the authoring contract (mirror existing cases at the tier →
draft via `npm run eval:mine` or by hand against `eval/cases/schema.json` →
`golden_pending: true` → scaffold `eval/goldens/<id>/` → validate schema + id/tier
match + `npx vitest run`). It will hand off with the next step: capture the golden
on the VM via the eval-implementer role. Show me the drafted case JSON.

Installed into a catalogue, chosen by a router

Install eval-author and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

Photo: RDNE Stock project

For the business

What it costs to adopt - in engineering time.

Same client, same code
One base URL
The OpenAI-compatible endpoint answers your existing SDK calls; model=auto lets the router choose.
The SDKs
One URL, every vendor
No vendor lock-in
Point your OpenAI-compatible client at one address; add, swap or retire providers in the catalogue without touching application code.
Every vendor side by side
Next eligible model answers
Failover built in
When a provider is down or over quota the request moves on; the trace says it did.

dynamics365ninja-d365fo-mcp-server-eval-author-command.prompt.jsonjson

{
  "prompt": "Launch the **eval-author** subagent (Task tool, `subagent_type: eval-author`) to\ndraft a new case for the eval catalog.\n\nWhat to author: **{{arguments}}**\n\nThe subagent carries the authoring contract (mirror existing cases at the tier →\ndraft via `npm run eval:mine` or by hand against `eval/cases/schema.json` →\n`golden_pending: true` → scaffold `eval/goldens/<id>/` → validate schema + id/tier\nmatch + `npx vitest run`). It will hand off with the next step: capture the golden\non the VM via the eval-implementer role. Show me the drafted case JSON.",
  "variables": [
    "arguments"
  ],
  "notes": "Draft a new eval case (eval/cases/<id>.json) + scaffold its golden folder, golden_pending until captured on the VM. Tools: Agent, Bash, Read, Edit, Write, Grep, Glob. Arguments: <feature description or failure to turn into a case>.",
  "metadata": {
    "source": {
      "provider": "github-claude-commands",
      "repository": "https://github.com/dynamics365ninja/d365fo-mcp-server",
      "path": ".claude/commands/eval-author.md",
      "ref": "355998507dc88bf0ef993d6f254321b5bcdb4eaa",
      "url": "https://github.com/dynamics365ninja/d365fo-mcp-server/blob/355998507dc88bf0ef993d6f254321b5bcdb4eaa/.claude/commands/eval-author.md",
      "key": "dynamics365ninja/d365fo-mcp-server/.claude/commands/eval-author.md"
    }
  }
}

Fetch it by URL: GET /api/v1/registry/dynamics365ninja-d365fo-mcp-server-eval-author-command/manifest?version=1.0.0

eval-author

About

Use it

Install eval-author and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Manifest

Reviews