repository·main·Indexed 40 days ago
https://github.com/limitless-ai-inc/limitless-api-examplesPractical, LLM-powered code examples for interacting with the Limitless Developer API. Includes implementations in Python and TypeScript for common use cases such as exporting transcript data as Markdown, generating daily summaries via OpenAI, and visualizing lifelog data usage trends.
To use these examples, install the dependencies and configure your environment variables.
Install dependencies using npm:
npm installCreate a .env file in the root directory with the following keys:
LIMITLESS_API_KEY: Your Limitless API key.OPENAI_API_KEY: Your OpenAI API key (required for summarization examples).LIMITLESS_API_URL: The API endpoint (defaults to https://api.limitless.ai).npm installAll API requests require authentication via an API key. You must include your key in the request header using the X-API-KEY key.
curl -X GET "https://api.limitless.ai/v1/<endpoint>" \
-H "X-API-KEY: YOUR_API_KEY"To set up the local Python environment for running the provided examples, navigate to the examples directory, create a virtual environment, and install the required dependencies.
cd examples
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txtTo use the examples in this repository, ensure you have the following prerequisites:
virtualenv installed.Alternatively, you can use the provided openapi.yml file to generate client libraries in your preferred programming language.
To run the lifelog visualization notebook, you need to install pandas, matplotlib, and requests via pip.
%pip install pandas matplotlib requestsThe summarize_day.py script generates a summary of your transcripts. This script requires both your LIMITLESS_API_KEY and an OPENAI_API_KEY for the LLM processing. The output will stream directly to the console in Markdown format.
LIMITLESS_API_KEY="your_api_key" OPENAI_API_KEY="sk-..." python summarize_day.pysummarize_day.ts script demonstrates how to fetch lifelogs and use OpenAI to generate a summary of the day's activities. By default, it fetches the 10 most recent lifelogs.export_markdown.ts script demonstrates how to fetch lifelogs from the API and export them into markdown format. By default, this script fetches only the most recent lifelog.The export_markdown.py script prints the Markdown content from the most recent record. It requires the LIMITLESS_API_KEY environment variable.
LIMITLESS_API_KEY="your_api_key" python export_markdown.py_client.ts. It provides a getLifelogs function which is used to fetch lifelog data. This function handles pagination and accepts various options to customize the fetch request.Use the following npm scripts to manage the project and run specific examples:
npm run build: Compiles the TypeScript source code into JavaScript.npm run export-markdown: Executes the export_markdown.ts example.npm run summarize-day: Executes the summarize_day.ts example.npm run build
npm run export-markdown
npm run summarize-dayYou can visualize your recording periods throughout the day by downloading lifelog data from the Limitless API and plotting the start and end timestamps using pandas and matplotlib. This process involves fetching lifelogs for a specific date, extracting the startTime from the contents array of each lifelog, and generating a timeline chart showing when recording sessions occurred.
import pandas as pd
import matplotlib.pyplot as plt
import requests
from datetime import date
# 1. Setup API Key
API_KEY = "< your api key >"
# 2. Define download function (simplified logic)
def download_lifelogs(date_str=None, page_size=10):
url = "https://api.limitless.ai/v1/lifelogs"
params = {
"date": date_str or date.today().isoformat(),
"timezone": "America/Chicago",
"includeMarkdown": "false",
"direction": "asc",
"limit": page_size,
"includeHeadings": "false"
}
headers = {"X-API-Key": API_KEY}
# ... (pagination logic using nextCursor) ...
return df
# 3. Download and Plot
df = download_lifelogs()
if not df.empty:
plt.figure(figsize=(12, 6))
for i, row in df.iterrows():
start_hour = row['first_timestamp'].hour + row['first_timestamp'].minute/60
end_hour = row['last_timestamp'].hour + row['last_timestamp'].minute/60
plt.plot([start_hour, end_hour], [i, i], linewidth=3, color='blue')
plt.show()