# //beforeyouship — LLM cost modeling from your editor

//beforeyouship — LLM cost modeling from your editor is an MCP server for Claude Code that estimates an LLM app’s monthly cost using architecture presets, token prices and three cost scenarios.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 169 installs
- Author: beforeyouship
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/beforeyouship/cost-model
- Updated: 2026-06-11
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/beforeyouship-llm-cost-modeling-from-your-editor

## What it does

Three tools land in your editor. `list_archetypes` hands back seven preset LLM architectures (chatbot, RAG pipeline, multi-step agent and others), each with default usage numbers. `get_model_prices` lists per-1M-token rates for input, output, cached input and batch, along with context windows and a note on how stale the pricing is. `estimate_cost` takes an archetype id plus `calls_per_day` and returns a monthly figure.

The model behind it accounts for retried calls, prompt caching, batch discounts, infra overhead and 3×/10× traffic growth. You get Naive, Realistic and Worst Case dollars per month for each model, growth scenarios, and a recommended pick. Token counts and multipliers can be overridden.

No key is needed to start: demo mode covers six free-tier models. A Pro key from beforeyouship.dev opens the 18-model catalog, so without it you compare across six models only — thin ground when the shortlist is GPT-5.x, Claude, Gemini and DeepSeek. Point your client at `https://cost-model--beforeyouship.run.tools`.

## Requirements
- Remote server: `https://cost-model--beforeyouship.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http cost-model https://cost-model--beforeyouship.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add beforeyouship-llm-cost-modeling-from-your-editor --url https://cost-model--beforeyouship.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "beforeyouship-llm-cost-modeling-from-your-editor": {
            "url": "https://cost-model--beforeyouship.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "beforeyouship-llm-cost-modeling-from-your-editor": {
            "type": "http",
            "url": "https://cost-model--beforeyouship.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http beforeyouship-llm-cost-modeling-from-your-editor https://cost-model--beforeyouship.run.tools
```

## When to use it
- model the monthly bill for an LLM app before writing code
- compare Claude, Gemini and GPT-5.x on your own traffic profile
- see what 3× and 10× growth does to the cost
- pull current per-1M-token rates including cached and batch pricing
- start from a preset RAG or agent archetype instead of a blank estimate

## Example prompts
- List the archetypes and estimate what a multi-step agent costs per month.
- Run `estimate_cost` for a chatbot on my traffic and compare Claude against Gemini.
- Pull `get_model_prices` and flag any pricing that has gone stale.

## Similar
- [cost-aware-llm-pipeline](https://affpapa.org/en/nevibecode/skills/cost-aware-llm-pipeline.md) — skill · 6400 installs
- [llm-cost-optimization](https://affpapa.org/en/nevibecode/skills/llm-cost-optimization.md) — skill · 370 installs
- [llm-cost-optimizer](https://affpapa.org/en/nevibecode/skills/llm-cost-optimizer.md) — skill · 73 installs
- [cost-report](https://affpapa.org/en/nevibecode/skills/cost-report.md) — skill · ★ 1198 repo stars
- [Mac Shortcuts — read your Mac from any agent](https://affpapa.org/en/nevibecode/mcp/mac-shortcuts-read-your-mac-from-any-agent.md) — MCP server · ★ 15 repo stars
- [ATOM Pricing Intelligence](https://affpapa.org/en/nevibecode/mcp/atom-pricing-intelligence.md) — MCP server · 4780 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
