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.
Prompt file imported from wololobzh/dedans (.github/prompts/review-feature.prompt.md). Copyright stays with the author.
Use the reviewer and qa agents to review the current feature/change.
Check domain correctness, authorization, auditability, metric semantics, DB access boundaries, regression risk and test coverage. Classify findings as blocking, important or optional.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
Use the reviewer and qa agents to review the current feature/change.
Check domain correctness, authorization, auditability, metric semantics, DB access boundaries, regression risk and test coverage. Classify findings as blocking, important or optional.
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
{
"prompt": "Use the reviewer and qa agents to review the current feature/change.\n\nCheck domain correctness, authorization, auditability, metric semantics, DB access boundaries, regression risk and test coverage. Classify findings as blocking, important or optional.",
"variables": [],
"notes": "Perform an adversarial final review of a feature.",
"metadata": {
"source": {
"provider": "github-prompts",
"repository": "https://github.com/wololobzh/dedans",
"path": ".github/prompts/review-feature.prompt.md",
"ref": "c6f010ba362d0036d2493c62a5bd946b81cf4120",
"url": "https://github.com/wololobzh/dedans/blob/c6f010ba362d0036d2493c62a5bd946b81cf4120/.github/prompts/review-feature.prompt.md",
"key": "wololobzh/dedans/.github/prompts/review-feature.prompt.md"
}
}
}
Installed into a catalogue, chosen by a router
Install review-feature 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.