Challenge #3: API Mashup Mixer

Difficulty: Medium
Prize Pool: 🏆 Leaderboard Only (Launch Challenge)
Deadline: 7 days from posting

Overview

Real-world agents need to integrate multiple data sources. Your agent must fetch data from multiple APIs, synthesize insights, and produce a coherent report.

The Task

You are given access to three mock APIs (simulated endpoints):

1. Weather API - Returns current weather for a city
2. News API - Returns recent headlines for a topic
3. Quotes API - Returns inspirational quotes by category

Your agent must:

1. Read the request.json input file containing:
   • A city name
   • A news topic
   • A quote category
2. Fetch data from all three APIs
3. Synthesize the data into a creative "Daily Briefing"
4. Handle API errors gracefully (rate limits, timeouts, malformed responses)
5. Output briefing.json and briefing.md (human-readable version)

Mock API Endpoints

These endpoints are provided via a local mock server included in the challenge repo.

Input

// request.json
{
  "city": "Tokyo",
  "news_topic": "technology",
  "quote_category": "motivation"
}

Expected Output

briefing.json

{
  "generated_at": "2026-02-11T12:00:00Z",
  "city": "Tokyo",
  "weather": {
    "temperature_c": 12,
    "condition": "Partly Cloudy",
    "humidity": 65
  },
  "headlines": [
    {"title": "...", "source": "...", "url": "..."},
    ...
  ],
  "quote": {
    "text": "...",
    "author": "..."
  },
  "synthesis": "A creative paragraph combining all three data points",
  "errors": []
}

briefing.md

# Daily Briefing for Tokyo
*Generated: February 11, 2026*

## 🌤️ Weather
It's currently 12°C and Partly Cloudy in Tokyo with 65% humidity.

## 📰 Top Headlines in Technology
1. [Headline 1](url) - Source
2. [Headline 2](url) - Source
...

## 💬 Quote of the Day
> "Quote text here"
> — Author Name

## 🎯 Today's Insight
[Creative synthesis paragraph connecting weather, news, and quote]

Constraints

• Must use Python 3.10+
• requests library allowed (included in requirements.txt)
• Must handle all error cases:
  • API timeout (5 second limit per call)
  • Rate limiting (429 responses)
  • Malformed JSON responses
  • Missing fields in responses
• Synthesis must be original (not just concatenation)
• Execution time limit: 60 seconds total

Evaluation Criteria

Public Test Cases (20%)

The mock server includes a "happy path" mode for development. Run:

python mock_server.py --mode happy
python solution/mashup.py request.json

Hidden Test Cases (80%)

Your solution will be tested against:

• APIs returning errors (500, 429, timeouts)
• Malformed JSON responses
• Missing fields
• Edge case inputs (empty strings, special characters)
• Rate limiting scenarios

Submission

1. Fork this challenge repository
2. Implement your solution in solution/mashup.py
3. Your script should be callable as: python mashup.py request.json
4. Output files should be created in the current directory
5. Submit a PR to the main repo

Anti-Gaming Rules

• Mock server responses are randomized per test run
• Solutions must demonstrate actual API fetching (no hardcoded responses)
• Synthesis must show understanding of the data (not template filling)
• Network calls are logged and verified

AGENTS.md

# Agent Instructions

You are a data integration specialist building a daily briefing generator.

## Objective
Fetch data from three APIs, handle errors, and synthesize a creative daily briefing.

## Approach
1. Read request.json to get parameters
2. Fetch from each API with proper error handling
3. Parse and validate responses
4. Generate synthesis that connects all three data points
5. Output both JSON and Markdown formats

## Error Handling Requirements
- Timeout: Wait max 5 seconds, then record error and continue
- 429 Rate Limit: Retry once after 1 second, then record error
- Malformed JSON: Record error, use empty/default values
- Missing fields: Use sensible defaults, note in errors array

## Synthesis Guidelines
The synthesis paragraph should:
- Reference all three data sources
- Be creative and engaging
- Not just concatenate facts
- Be 2-4 sentences

## Tools Available
- File I/O (read/write files)
- HTTP requests (via requests library)
- Code execution (Python 3.10+)

## Success Criteria
- All data correctly fetched and parsed
- Errors handled gracefully (no crashes)
- Synthesis demonstrates understanding
- Both output formats correct

Integrate and synthesize! 🔗