Slice · AI prompt

AI Router — Backend Provider Proxy

Backend infrastructure (no UI). Single proxy that every other ai-* feature calls. Tier-routed — nano (Haiku) for classification, mid (Sonnet) for chat, flagship (Opus) for deep reasoning. Per-call usage log + cost guard. Works with direct provider keys or OpenRouter umbrella. Not something you mount — installed automatically as a peer when you add ai-workspace.

Resource detail View source

Copy this prompt into Claude / Codex / Cursor

ai-router.prompt.mdmarkdown

# Install `ai-router` — AI Router — Backend Provider Proxy

> Backend infrastructure (no UI).

📚 Knowledge base : https://resource.rahmanef.com/llms.txt
📦 Slice detail   : https://resource.rahmanef.com/slices/ai-router
🧠 JSON catalog   : https://resource.rahmanef.com/api/knowledge?slice=ai-router
🔗 Source         : https://github.com/rahmanef63/resource-site/tree/main/frontend/slices/ai-router

## 1. Install

```bash
npm i ai @openrouter/ai-sdk-provider
# alias: npx rr add ai-router
```

The CLI copies `frontend/slices/ai-router/` into your project + augments `.env.example` + installs npm deps automatically. Run it from your project root.

## 2. What it ships

- npm: `ai`, `@openrouter/ai-sdk-provider`
- shadcn primitives: `button`
- env vars:
  - `OPENROUTER_API_KEY` (convex, required)
- convex feature paths: `convex/features/ai`

## 3. Wire it up

Run `npx rr add ai-router`. Wrap every AI call through ai-router. Tiers: nano = quick classification (spam-flag, headline-suggest), mid = chat / draft, flagship = methodology-review / deep-think. Token usage logs to ai_usage table for the cost dashboard.

## Rules of engagement

- shadcn-only UI primitives. No raw `<button>` / `<dialog>` / native date or file inputs.
- 200-line hard cap per source file (extract neighbours when over).
- All Convex queries hit an index (`.withIndex(...)`); never bare `.collect()`.
- Public mutations/queries declare `args:` validators + authz.
- Full ruleset: https://resource.rahmanef.com/best-practice

The agent will fetch /llms.txt for the full ruleset and use /api/knowledge for the JSON catalog.

AI Router — Backend Provider Proxy

# Install `ai-router` — AI Router — Backend Provider Proxy > Backend infrastructure (no UI). 📚 Knowledge base : https://resource.rahmanef.com/llms.txt 📦 Slice detail : https://resource.rahmanef.com/slices/ai-router 🧠 JSON catalog : https://resource.rahmanef.com/api/knowledge?slice=ai-router 🔗 Source : https://github.com/rahmanef63/resource-site/tree/main/frontend/slices/ai-router ## 1. Install ```bash npm i ai @openrouter/ai-sdk-provider # alias: npx rr add ai-router ``` The CLI copies `frontend/slices/ai-router/` into your project + augments `.env.example` + installs npm deps automatically. Run it from your project root. ## 2. What it ships - npm: `ai`, `@openrouter/ai-sdk-provider` - shadcn primitives: `button` - env vars: - `OPENROUTER_API_KEY` (convex, required) - convex feature paths: `convex/features/ai` ## 3. Wire it up Run `npx rr add ai-router`. Wrap every AI call through ai-router. Tiers: nano = quick classification (spam-flag, headline-suggest), mid = chat / draft, flagship = methodology-review / deep-think. Token usage logs to ai_usage table for the cost dashboard. ## Rules of engagement - shadcn-only UI primitives. No raw `<button>` / `<dialog>` / native date or file inputs. - 200-line hard cap per source file (extract neighbours when over). - All Convex queries hit an index (`.withIndex(...)`); never bare `.collect()`. - Public mutations/queries declare `args:` validators + authz. - Full ruleset: https://resource.rahmanef.com/best-practice