Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
README - Prompt - OpenSmartRoute
Promptv1.0.0
README
Cursor runs delivery via the workflow entry files, not via duplicated local agent folders.
Prompt file imported from meta-secret/meta-secret-core (.cursor/commands/README.md). Copyright stays with the author.
Cursor commands (workflow entry)
Cursor runs delivery via the workflow entry files, not via duplicated local agent folders.
Single source of truth:
.ai/WORKFLOW.md
.ai/PIPELINE.md
Use in Cursor chat:
run issue <id-or-text>
run issue <id-or-text> --from stage-<n>
Issue intake in this repo is GitHub-only (gh) through github-issue-coordinator.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
# Cursor commands (workflow entry)
Cursor runs delivery via the workflow entry files, not via duplicated local agent folders.
Single source of truth:
- `.ai/WORKFLOW.md`
- `.ai/PIPELINE.md`
Use in Cursor chat:
- `run issue <id-or-text>`
- `run issue <id-or-text> --from stage-<n>`
Issue intake in this repo is GitHub-only (`gh`) through `github-issue-coordinator`.
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
Installed into a catalogue, chosen by a router
Install README 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.