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 Ryukijano/agent-skills (.cursor/commands/ai-for-communication.md). Copyright stays with the author.
ai-for-communication
Quick reference for ai for communication.
Architecture / framework: ask first.
Key command / env vars: see the full skill.
Verification: run the smallest failing test first.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
# ai-for-communication
Quick reference for`ai for communication`.
- Architecture / framework: ask first.
- Key command / env vars: see the full skill.
- Verification: run the smallest failing test first.
Manifest
The prompt text and its fill-in variables. Copy it or fetch it by URL from your own code.
{
"prompt": "# ai-for-communication\n\nQuick reference for `ai for communication`.\n\n- Architecture / framework: ask first.\n- Key command / env vars: see the full skill.\n- Verification: run the smallest failing test first.",
"variables": [],
"notes": "Quick reference for `ai for communication`.",
"metadata": {
"source": {
"provider": "github-cursor-commands",
"repository": "https://github.com/Ryukijano/agent-skills",
"path": ".cursor/commands/ai-for-communication.md",
"ref": "80e9045f2ab7a2549b5e058368f0706e690396c1",
"url": "https://github.com/Ryukijano/agent-skills/blob/80e9045f2ab7a2549b5e058368f0706e690396c1/.cursor/commands/ai-for-communication.md",
"key": "Ryukijano/agent-skills/.cursor/commands/ai-for-communication.md"
}
}
}
Installed into a catalogue, chosen by a router
Install ai-for-communication 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.