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.
Load the ck-plan-json skill from .cursor/skills/ck-plan-json/SKILL.md
Parse the user's feature request
Generate plan.json in plans/<slug>/
Validate the plan structure
Output next step: /cook --json plans/<slug>/plan.json
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
`/plan` — Create implementation plan
1. Load the `ck-plan-json` skill from`.cursor/skills/ck-plan-json/SKILL.md`2. Parse the user's feature request
3. Generate `plan.json`in`plans/<slug>/`4. Validate the plan structure
5. Output next step: `/cook --json plans/<slug>/plan.json`
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
{
"prompt": "`/plan` — Create implementation plan\n\n1. Load the `ck-plan-json` skill from `.cursor/skills/ck-plan-json/SKILL.md`\n2. Parse the user's feature request\n3. Generate `plan.json` in `plans/<slug>/`\n4. Validate the plan structure\n5. Output next step: `/cook --json plans/<slug>/plan.json`",
"variables": [],
"notes": "`/plan` — Create implementation plan",
"metadata": {
"source": {
"provider": "github-cursor-commands",
"repository": "https://github.com/huuanh20/awesome-ai-agent-skills",
"path": ".cursor/commands/plan.md",
"ref": "659b7a01f6a83746017994abaffd6ecf9738a6af",
"url": "https://github.com/huuanh20/awesome-ai-agent-skills/blob/659b7a01f6a83746017994abaffd6ecf9738a6af/.cursor/commands/plan.md",
"key": "huuanh20/awesome-ai-agent-skills/.cursor/commands/plan.md"
}
}
}
Installed into a catalogue, chosen by a router
Install plan 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.