One SDK for Email, Telegram, Discord, Slack, and Google Sheets — and the execution backend your AI agents need. Write automation logic in Python — we run it in the cloud.
from apider import Email, Telegram, CloudScheduler def monitor_inbox(): emails = Email.read(unread_only=True) if emails: summary = "\n".join([f"• {e.subject}" for e in emails]) Telegram.send(f"📬 {len(emails)} new emails:\n{summary}") # Runs every 10 minutes — in the cloud, forever CloudScheduler.every_minutes(10, monitor_inbox)
from apider import Email # Send Email.send( to="team@company.com", subject="Daily digest", body="Here's your report 📊", attachments=["report.pdf"] ) # Read emails = Email.read(unread_only=True, mark_as_read=True) for e in emails: print(e.subject, e.email_sender)
from apider import Sheets # Read as Polars DataFrame with type inference df = Sheets.read(filters={"status": "pending"}) # Append a new row Sheets.append(["2026-04-23", "Jorge", 1499]) # Upsert by column match Sheets.upsert("email", "user@co.com", ["user@co.com", "Pro"])
Stop writing 100 lines to send an email. Stop wrestling with OAuth. Just import and run.
Email.send() · Telegram.send() · Sheets.append(). No setup wizards, no OAuth flows, no SDK configuration.
Your scripts run on Azure Functions. No servers to provision, no containers to manage, no Dockerfile.
CloudScheduler.every_minutes(10, fn) registers a persistent job in the cloud. No crontab, no Celery.
Your API keys stay in .env on your machine. We never store them — they travel encrypted per-request.
Sheets.read() returns a Polars DataFrame with auto-detected types — integers, floats, strings. No more raw text.
Mix Email + Telegram + Discord + Slack in the same script. One execution, multiple outputs.
Apider exposes all its modules as MCP tools. Claude, Cursor, or any AI agent can send emails, read Sheets, and call any API — without a local server running.
Most MCP servers run on your machine — which means your laptop must stay open for your AI agent to work. Apider runs in Azure. Your agent executes tasks 24/7, even when you're offline.
{ "mcpServers": { "apider": { "command": "apider", "args": ["mcp-proxy"], "env": { "APIDER_ENDPOINT": "https://...", "APIDER_FUNCTION_KEY": "<key>", "APIDER_ENCRYPTION_KEY": "<key>" } } } }
No infrastructure knowledge required. If you can write a function, you can automate anything.
pip install apider. Add your API keys to .env. That's it — your credentials never leave your machine.
Use Email, Telegram, Sheets, Discord, Slack — all as simple Python functions. No boilerplate, no classes to extend.
CloudScheduler.every_minutes(10, fn) deploys your script to Azure. It runs forever, even when your laptop is off.
Pay only for what you use. No hidden fees, no credit card required to start.
Install Apider, write your first script, and run it in the cloud — in under 5 minutes.