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 deftai/directive (.codex/prompts/deft-directive-change-archive.md). Fill in {{arguments}} before use. Copyright stays with the author.
Read and follow node_modules/@deftai/directive-runtime/commands.md.
Honor {{arguments}} as documented for this command.
Do not inline the strategy, skill, or commands.md body here.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
Read and follow `node_modules/@deftai/directive-runtime/commands.md`.
Honor `{{arguments}}`as documented for this command.
Do not inline the strategy, skill, or commands.md body here.
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
{
"prompt": "Read and follow `node_modules/@deftai/directive-runtime/commands.md`.\nHonor `{{arguments}}` as documented for this command.\nDo not inline the strategy, skill, or commands.md body here.",
"variables": [
"arguments"
],
"notes": "Archive the completed change under history/archive/",
"metadata": {
"source": {
"provider": "github-codex-prompts",
"repository": "https://github.com/deftai/directive",
"path": ".codex/prompts/deft-directive-change-archive.md",
"ref": "f3696987bedaac39796adc17e280a4a758c49bc0",
"url": "https://github.com/deftai/directive/blob/f3696987bedaac39796adc17e280a4a758c49bc0/.codex/prompts/deft-directive-change-archive.md",
"key": "deftai/directive/.codex/prompts/deft-directive-change-archive.md"
}
}
}
Installed into a catalogue, chosen by a router
Install deft-directive-change-archive 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.