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

schema - Prompt - OpenSmartRoute

Promptv1.0.0

schema

Name: schema
Author: vastuhlfauth

Document a raw dataset in docs/data before any step uses it

by vastuhlfauth(0) 0 installs

Free

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

See reviews

About

Prompt file imported from vastuhlfauth/phd_model (.github/prompts/schema.prompt.md). Fill in {{dataset}} before use. Copyright stays with the author.

Document the dataset in data/raw/{{dataset}}. Do not modify the data.

List the files, including the content of any zip archive.
For each table file, use DuckDB to get the schema, the row count and 20 sample rows, without loading the whole file into memory:
- CSV: read_csv;
- Parquet: read_parquet;
- GeoPackage, Shapefile or GeoJSON: ST_Read;
- OSM PBF: ST_ReadOSM, limited to a few thousand rows.
Write docs/data/{{dataset}}.md with the template of docs/data/README.md:
- source and vintage, files, row counts;
- for each column: name, type, unit, meaning, example;
- keys and joins with other datasets, CRS;
- quirks: missing values, special codes, encodings, separators.
Add one line for the dataset in the index of docs/data/README.md.

Use it

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

Document the dataset in `data/raw/{{dataset}}`. Do not modify the data.

1. List the files, including the content of any zip archive.
2. For each table file, use DuckDB to get the schema, the row count and 20 sample rows, without loading the whole file into memory:
   - CSV: `read_csv`;
   - Parquet: `read_parquet`;
   - GeoPackage, Shapefile or GeoJSON: `ST_Read`;
   - OSM PBF: `ST_ReadOSM`, limited to a few thousand rows.
3. Write `docs/data/{{dataset}}.md` with the template of [docs/data/README.md](../../docs/data/README.md):
   - source and vintage, files, row counts;
   - for each column: name, type, unit, meaning, example;
   - keys and joins with other datasets, CRS;
   - quirks: missing values, special codes, encodings, separators.
4. Add one line for the dataset in the index of `docs/data/README.md`.

Installed into a catalogue, chosen by a router

Install schema 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.

vastuhlfauth-phd-model-schema-prompt.prompt.jsonjson

{
  "prompt": "Document the dataset in `data/raw/{{dataset}}`. Do not modify the data.\n\n1. List the files, including the content of any zip archive.\n2. For each table file, use DuckDB to get the schema, the row count and 20 sample rows, without loading the whole file into memory:\n   - CSV: `read_csv`;\n   - Parquet: `read_parquet`;\n   - GeoPackage, Shapefile or GeoJSON: `ST_Read`;\n   - OSM PBF: `ST_ReadOSM`, limited to a few thousand rows.\n3. Write `docs/data/{{dataset}}.md` with the template of [docs/data/README.md](../../docs/data/README.md):\n   - source and vintage, files, row counts;\n   - for each column: name, type, unit, meaning, example;\n   - keys and joins with other datasets, CRS;\n   - quirks: missing values, special codes, encodings, separators.\n4. Add one line for the dataset in the index of `docs/data/README.md`.",
  "variables": [
    "dataset"
  ],
  "notes": "Document a raw dataset in docs/data before any step uses it Arguments: dataset folder name under data/raw, e.g. filosofi.",
  "metadata": {
    "source": {
      "provider": "github-prompts",
      "repository": "https://github.com/vastuhlfauth/phd_model",
      "path": ".github/prompts/schema.prompt.md",
      "ref": "9a1dc52f9fd27e5a007dae13d7ac435272d309c8",
      "url": "https://github.com/vastuhlfauth/phd_model/blob/9a1dc52f9fd27e5a007dae13d7ac435272d309c8/.github/prompts/schema.prompt.md",
      "key": "vastuhlfauth/phd_model/.github/prompts/schema.prompt.md"
    }
  }
}

Fetch it by URL: GET /api/v1/registry/vastuhlfauth-phd-model-schema-prompt/manifest?version=1.0.0

About

Use it

Install schema 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