PRarena Data Pipeline Runner

This skill automates the complete workflow for the PRarena project, which tracks opened and merged PRs by top SWE coding agents from OpenAI, GitHub, and others.

What This Skill Does

Runs the three-phase PRarena pipeline:

1. Collects fresh PR data into `data.csv`

2. Generates visualization charts and renders templates (index, README)

3. Starts a local web server to preview results

Instructions

When the user requests to run the PRarena pipeline or update PR tracking data, follow these steps:

Step 1: Activate Virtual Environment

Always use the project's virtual environment named `venv`:

```bash

source venv/bin/activate

```

On Windows:

```bash

venv\Scripts\activate

```

Step 2: Collect Data

Run the data collection script to fetch new PR data:

```bash

python3 collect_data.py

```

**What this does:**

Step 3: Generate Charts and Templates

Process the collected data and generate visualizations:

```bash

python3 generate_chart.py

```

**What this does:**

Step 4: Start Local Web Server

Launch a local server to preview the results:

```bash

python3 -m http.server 8000

```

**What this does:**

Full End-to-End Command Sequence

```bash

source venv/bin/activate

python3 collect_data.py

python3 generate_chart.py

python3 -m http.server 8000

```

Usage Examples

**User:** "Update the PRarena data and show me the results"

**Action:** Run all three scripts in sequence, then provide the localhost URL

**User:** "Regenerate the charts without fetching new data"

**Action:** Skip `collect_data.py`, run only `generate_chart.py` and the web server

**User:** "Just collect the latest data"

**Action:** Run only `collect_data.py` with venv activated

Important Notes