Skip to main content
Glama

agentforge-mcp

Stop re-explaining what you want to your AI coding agent. Hand it one rough sentence — get back a structured, tool-tuned prompt it can execute in one pass.

npm license

agentforge.sciscale.org · Quick start · npm

AgentForge MCP in action

agentforge-mcp is the Model Context Protocol server for AgentForge. One tool, one job: turn a vague request into a sharp one.

Why

You know the loop: you ask your coding agent for something, it misreads the half you didn't spell out, you correct it, it breaks something else, you re-explain. The fix isn't a smarter agent — it's a sharper prompt.

AgentForge does the prompt engineering for you. Give it "add a dark mode toggle that persists" and it:

  • extracts the real requirements — the edge cases, the acceptance criteria, the things you'd have forgotten to mention;

  • formats for your specific agent — Claude Code wants file:line references, Cursor wants directives, Aider wants tight context;

  • quality-checks before returning — every prompt is scored across 12 dimensions before it leaves the engine.

Harness engineering is having a moment — but prompt engineering matters more in the coding-agent era, not less. Tested across 1,000+ real coding cases, projects with AgentForge in the loop reached a deployable state 147% faster.

Related MCP server: MCP Vibe Coding Tools

Quick start

No install — it runs through npx.

1. Get an API key. Sign in at agentforge.sciscale.org, open API keys, and create one. The key is shown once — copy it.

2. Add it to your agent.

Claude Code:

claude mcp add agentforge --env AGENTFORGE_API_KEY=af_your_key -- npx -y agentforge-mcp

Cursor / Windsurf / Claude Desktop — add to your MCP config (~/.cursor/mcp.json, ~/.codeium/windsurf/mcp_config.json, etc.):

{
  "mcpServers": {
    "agentforge": {
      "command": "npx",
      "args": ["-y", "agentforge-mcp"],
      "env": { "AGENTFORGE_API_KEY": "af_your_key" }
    }
  }
}

3. Use it. Ask your agent naturally — "refine this with AgentForge, then build it: add a dark mode toggle that persists."

The tool — agentforge_refine_prompt

Argument

Default

request

—

Your task in plain language (1–4000 chars). Rough is fine.

target_tool

claude-code

claude-code, codex, cursor, aider, continue, windsurf, kimi, generic

style

plan-first

plan-first, direct-edit, explore-first

Returns the refined prompt, plus its Quality Engine score and your remaining daily usage.

Free vs Pro

Free

Pro

Refinements

3 / day

Unlimited

Quality Engine

scored across 12 dimensions

scored + auto-refined until it passes ≥ 90

Or use the web app

Don't want to wire up an MCP server at all? The same engine — same Quality Engine, same per-account Pro — runs right in your browser at agentforge.sciscale.org. No install, no key, no config.

The AgentForge web app

→ Try it now at agentforge.sciscale.org

Paste a request, pick your target tool, copy the prompt — generation history and advanced modes included:

The web app in action

How it works

agentforge-mcp is a thin client — no engine logic ships in this package. Your request goes to the AgentForge API, the hosted engine does the extraction, formatting, and quality-checking, and the prompt comes back. The engine keeps improving without you ever updating this package.

Configuration

Variable

Default

AGENTFORGE_API_KEY

—

Required. Your API key.

AGENTFORGE_API_URL

https://agentforge.sciscale.org/api/v1/refine

Override the endpoint (rarely needed).

Develop

npm install
npm run build      # tsc -> dist/
node dist/index.js # runs on stdio

License

MIT — see LICENSE.

agentforge-mcp is part of AgentForge — a SciScale studio product.

Available Tools

1 tool
agentforge_refine_promptRefine a prompt with AgentForgeA

Turn a rough, plain-language coding request into a structured, tool-tuned prompt for an AI coding agent.

AgentForge extracts the real requirements, fills in missing edge cases, formats the prompt for the target agent's conventions, and runs it through a Quality Engine before returning it. Reach for this when a request is vague or under-specified and you want a sharper prompt before handing it to a coding agent.

Args:

  • request (string, required): the coding task in plain language, 1-4000 chars.

  • target_tool (string): one of claude-code, codex, cursor, aider, continue, windsurf, kimi, generic. Default: claude-code.

  • style (string): plan-first, direct-edit, or explore-first. Default: plan-first.

Returns the refined prompt as text. structuredContent additionally carries: { "prompt": string, "tier": "free" | "pro", "quality": { "score": number, "max_score": number, "passed": boolean } | null, "usage": { "used_today": number, "limit": number | null } }

Errors are returned as text, e.g.:

  • "API key not configured" — set AGENTFORGE_API_KEY.

  • "invalid or revoked API key" (401) — check the key at agentforge.sciscale.org.

  • "daily free limit reached" (429) — upgrade to Pro for unlimited calls.

Inputs: request (required), target_tool, style. Returns: prompt, tier, quality, usage

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.4/5.0

    Scored across 1 tool

    Disambiguation5/5

    With only one tool, there is no possibility of confusion between tools. The single tool has a clear, distinct purpose.

    Naming Consistency5/5

    The single tool name 'agentforge_refine_prompt' follows a consistent verb_noun pattern with a server prefix. As there is only one tool, naming consistency is perfect.

    Tool Count3/5

    Having a single tool is appropriate for the focused purpose of prompt refinement, but it falls on the low end of the typical well-scoped range (3-15 tools). The server could potentially benefit from a few related tools, but the current count is not problematic.

    Completeness5/5

    The tool fully covers its domain: it refines a coding request with parameters for target tool and style, and returns the refined prompt along with quality and usage information. There are no obvious missing features for its stated purpose.

    Maintenance

    ActivityInactive
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers