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

skill-evaluate - Prompt - OpenSmartRoute

Promptv1.0.0

skill-evaluate

Name: skill-evaluate
Author: gjb0660

Evaluate a concrete request via skill-reviewer Evaluation Mode.

by gjb0660(0) 0 installs

Free

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

See reviews

About

Prompt file imported from gjb0660/skills (.github/prompts/skill-evaluate.prompt.md). Copyright stays with the author.

Related skill: skill-reviewer. Load and follow Evaluation Mode section for template and principles.

Arguments (MUST VALIDATE):

evaluation request to execute explicitly

If the argument is missing or ambiguous, ask for the evaluation request via built-in question tool.

Delegate the evaluation workflow, comparison protocol, scoring assertions, and token accounting to skill-reviewer.

Prompt-specific rules:

State that the run is using Evaluation Mode.
Do not switch to Inspect Mode unless the user explicitly redirects you.
Do not restate or reimplement skill-reviewer's detailed evaluation protocol inside this prompt.
Keep the prompt layer minimal and let the skill own the heavy logic.

Remember to follow the skill-reviewer procedures in Evaluation Mode section.

Use it

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

Related skill: `skill-reviewer`. Load and follow **Evaluation Mode** section for template and principles.

Arguments (MUST VALIDATE):

* evaluation request to execute explicitly

If the argument is missing or ambiguous, ask for the evaluation request via built-in question tool.

Delegate the evaluation workflow, comparison protocol, scoring assertions, and token accounting to skill-reviewer.

Prompt-specific rules:

1. State that the run is using Evaluation Mode.
2. Do not switch to Inspect Mode unless the user explicitly redirects you.
3. Do not restate or reimplement skill-reviewer's detailed evaluation protocol inside this prompt.
4. Keep the prompt layer minimal and let the skill own the heavy logic.

Remember to follow the `skill-reviewer` procedures in **Evaluation Mode** section.

Installed into a catalogue, chosen by a router

Install skill-evaluate 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.

Manifest

The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.

gjb0660-skills-skill-evaluate-prompt.prompt.jsonjson

{
  "prompt": "Related skill: `skill-reviewer`. Load and follow **Evaluation Mode** section for template and principles.\n\nArguments (MUST VALIDATE):\n\n* evaluation request to execute explicitly\n\nIf the argument is missing or ambiguous, ask for the evaluation request via built-in question tool.\n\nDelegate the evaluation workflow, comparison protocol, scoring assertions, and token accounting to skill-reviewer.\n\nPrompt-specific rules:\n\n1. State that the run is using Evaluation Mode.\n2. Do not switch to Inspect Mode unless the user explicitly redirects you.\n3. Do not restate or reimplement skill-reviewer's detailed evaluation protocol inside this prompt.\n4. Keep the prompt layer minimal and let the skill own the heavy logic.\n\nRemember to follow the `skill-reviewer` procedures in **Evaluation Mode** section.",
  "variables": [],
  "notes": "Evaluate a concrete request via skill-reviewer Evaluation Mode. Arguments: What request should be evaluated?.",
  "metadata": {
    "source": {
      "provider": "github-prompts",
      "repository": "https://github.com/gjb0660/skills",
      "path": ".github/prompts/skill-evaluate.prompt.md",
      "ref": "23c63e62aa7b9c9bd925e14fb8eaa08ab72014b5",
      "url": "https://github.com/gjb0660/skills/blob/23c63e62aa7b9c9bd925e14fb8eaa08ab72014b5/.github/prompts/skill-evaluate.prompt.md",
      "key": "gjb0660/skills/.github/prompts/skill-evaluate.prompt.md"
    }
  }
}

Fetch it by URL: GET /api/v1/registry/gjb0660-skills-skill-evaluate-prompt/manifest?version=1.0.0

About

Use it

Install skill-evaluate 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