Attention Scalper

An AI Agent Skill built for the SagaPad Colosseum Hackathon Track.

The Attention Scalper is an execution bot designed to generate deterministic, 4-week X (Twitter) growth playbooks for early-stage crypto and hackathon projects. It operates by treating social media attention as a quantifiable market, producing highly-engineered content strategies.

Overview

This repository contains the backend Express API wrapper that powers the Attention Scalper Skill. It takes in project parameters (Tech Stack, One-Liner, Target Audience) and interfaces with OpenRouter to output a structured, multi-phase growth playbook.

Topic Addressed: Topic 1 — Hackathon Project Social Playbook Skill

Features

• Algorithmic Routing: Automatically determines the correct narrative tone based on a project's tech stack.
• Structured JSON Output: Returns clean, parseable JSON instantly consumable by any frontend.
• Enterprise Security: Built-in rate limiting, HTTP header protection (helmet), and strict payload validation to prevent prompt injection and DDOS.
• Docker Ready: Includes a Dockerfile for instant, one-click deployment to Railway, Render, or AWS.
• 4-Week Scalping Calendar: Generates a daily content schedule for the first month.
• The Genesis Thread: Writes the exact, word-for-word copy for a project's initial 5 launch posts.

Installation & Setup

1. Clone the repository:

   git clone https://github.com/belalangeth/attention-scalper.git
   cd attention-scalper
   

2. Install dependencies:

   npm install
   

3. Configure Environment Variables: Copy the example file and add your OpenRouter API Key.

   cp .env.example .env
   

4. Run the development server:

   npm run dev
   

Docker Deployment

To deploy using Docker, simply build and run the image:

docker build -t attention-scalper .
docker run -p 3000:3000 --env-file .env attention-scalper

API Usage

The skill exposes a single POST endpoint to generate the playbook.

Endpoint: `/api/generate-playbook` Method: `POST` Payload:

{
  "project_name": "Theta Scalper",
  "one_liner": "An automated high-frequency trading bot analyzing the first 120 seconds of a market window.",
  "tech_stack": "Python, Linux, Smart Contracts",
  "target_audience": "Algorithmic bot operators, high-frequency traders"
}

Example Output (JSON)

The Engine outputs strict, frontend-ready JSON without AI slop:

{
  "success": true,
  "data": {
    "story_angle": "Prediction markets are bleeding money to sentiment traders...",
    "four_week_calendar": [
      {
        "week": 1,
        "focus": "Deploy raw backtest data and architecture flexing."
      }
    ],
    "engagement_protocol": "Camp out in core MEV searcher replies. Quote tweet...",
    "genesis_posts": [
      "The market stops being deterministic exactly 120 seconds into a micro-trend...",
      "Standard RPC latency kills alpha before you even see the chart move..."
    ]
  }
}

Built With

• Node.js & Express
• TypeScript
• openai (OpenRouter Integration)

SagaPad Skill Link

[Insert your published SagaPad Skill link here]