# AgentForge

AgentForge turns a rough task description into a structured prompt for a selected coding agent. Use it when your request needs clearer requirements and acceptance criteria.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 2
- Author: sciscale-studio
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/sciscale-studio/agentforge-mcp
- Updated: 2026-05-28
- HTML: https://affpapa.org/en/nevibecode/mcp/agentforge

## What it does

The server sends a task description to the AgentForge API. Its hosted engine identifies requirements and edge cases, then shapes the prompt for the selected tool; the response also carries a quality score. Choose a plan-first, direct-edit, or explore-first style.

This package is a thin client, so prompt processing happens through the API. You need an `AGENTFORGE_API_KEY` to make requests; without it, the engine cannot be reached. Set `target_tool` to options such as `claude-code`, `codex`, or `cursor`. The free tier allows 3 refinements per day.

## Requirements
- Needs: Node.js (commands run via npx)
- Environment variables: `AGENTFORGE_API_KEY`, `AGENTFORGE_API_URL` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add --env AGENTFORGE_API_KEY=<AGENTFORGE_API_KEY> --env AGENTFORGE_API_URL=<AGENTFORGE_API_URL> --transport stdio agentforge -- npx -y agentforge-mcp
```
Replace <…> with your own values

### Codex

**Platform CLI**
```bash
codex mcp add agentforge --env AGENTFORGE_API_KEY=<AGENTFORGE_API_KEY> --env AGENTFORGE_API_URL=<AGENTFORGE_API_URL> -- npx -y agentforge-mcp
```
Replace <…> with your own values

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "agentforge": {
            "command": "npx",
            "args": [
                "-y",
                "agentforge-mcp"
            ],
            "env": {
                "AGENTFORGE_API_KEY": "<AGENTFORGE_API_KEY>",
                "AGENTFORGE_API_URL": "<AGENTFORGE_API_URL>"
            }
        }
    }
}
```
Replace <…> with your own values

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "agentforge": {
            "type": "stdio",
            "command": "npx",
            "args": [
                "-y",
                "agentforge-mcp"
            ],
            "env": {
                "AGENTFORGE_API_KEY": "<AGENTFORGE_API_KEY>",
                "AGENTFORGE_API_URL": "<AGENTFORGE_API_URL>"
            }
        }
    }
}
```
Replace <…> with your own values

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add -e AGENTFORGE_API_KEY=<AGENTFORGE_API_KEY> -e AGENTFORGE_API_URL=<AGENTFORGE_API_URL> agentforge npx -y agentforge-mcp
```
Replace <…> with your own values

## When to use it
- turn a brief task into a detailed agent prompt
- add acceptance criteria and edge cases to a request
- adapt instructions for a chosen coding agent
- check prompt quality before handing it to an agent

## Example prompts
- Refine this request with AgentForge for Claude Code: add a dark theme that remembers the user's choice.
- Create an AgentForge prompt for Codex from my task description.
- Rewrite my request for Cursor using a plan-first approach, then return the prompt.

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 392262
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33989
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81163
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119231
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47577
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43902

---
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