scalekit

A real working agent you can deploy

This repo uses a single SCALEKIT_USER_ID env var to simulate one user. In production, pass each user's real ID as the identifier on every Scalekit call, and send them an authorization link whenever their connector status is not ACTIVE. HubSpot's 30-minute token expiry is handled automatically.

01

Authorize, then orchestrate

main.py

Every connection is authorized once via a one-time link. Scalekit refreshes tokens for you, including HubSpot's 30-minute expiry. main.py triggers on a completed Granola transcript, verifies all four connectors are ACTIVE, then runs the post-call pipeline in sequence.

main.py

Copy

import os
from scalekit import ScalekitClient
from langchain.agents import create_tool_calling_agent

client = ScalekitClient(
    env_url=os.environ["SCALEKIT_ENV_URL"],
    client_id=os.environ["SCALEKIT_CLIENT_ID"],
    client_secret=os.environ["SCALEKIT_CLIENT_SECRET"],
)

# LangChain-compatible tools scoped to this user
tools = client.actions.langchain.get_tools(
    identifier="user@example.com",
    connection_names=["granolamcp", "hubspot", "gmail", "slack"],
)

agent = create_tool_calling_agent(llm, tools, prompt)

02

Read call transcript from Granola

read_transcript.py

03

Extract deal data with LLM

extract_data.py

04

Update HubSpot deal records

update_hubspot.py

05

Draft follow-up email and notify Slack

deliver_outputs.py

Why choose Scalekit

Delegated identity. Not service accounts.

Credentials never touch agent code or LLM context. The agent acts as the user, not as a shared bot.

Delegated OAuth - Agent reads your calendar, your inbox — scoped to the authorizing identity, not org-wide.

Credentials outside agent runtime - Tokens never touch agent code or LLM context. Both failure modes covered.

Token lifecycle automatic - Refresh, expiry, rotation across all connectors. One SDK call. Zero management code.

200+ prebuilt connectors - Google, Slack, HubSpot, GitHub, Jira, Notion, Salesforce — same auth pattern everywhere.

Agents that update your CRM, without the auth plumbing

Four things you'd otherwise build: Granola MCP auth, HubSpot OAuth with 30-minute refresh, Gmail credentials, Slack auth. Handled.

OAuth flow per connector

One SDK call returns a delegated token for any connector. Google, HubSpot, Slack, same pattern across all 200+ connectors

tools = client.actions.langchain.get_tools(
    identifier=user_id,
    connection_names=["granolamcp", "hubspot", "gmail"],
)
agent = create_tool_calling_agent(llm, tools, prompt)

Secure token vault

Scalekit stores OAuth credentials outside agent code and outside LLM context. Both are separate failure modes. Both covered

client = ScalekitClient(
    env_url=os.environ["SCALEKIT_ENV_URL"],
    client_id=os.environ["SCALEKIT_CLIENT_ID"],
    client_secret=os.environ["SCALEKIT_CLIENT_SECRET"],
)
# Credentials never in agent code or LLM context

Token refresh logic

Token lifecycle handled automatically — expiry, rotation, re-auth — across every connector. Agent runs in 6 months. Same call works

# Day 1 or day 180 — same call works
tools = client.actions.langchain.get_tools(
    identifier=user_id,
    connection_names=["granolamcp", "hubspot", "gmail"],
)

Try other Agent Templates

Prebuilt agents you can ship today

Each one runs on delegated identity, scoped per user.

SALES

Sales call prep agent

Pull Granola notes and Attio contact history to draft a pre-call brief before every sales meeting. Zero rep input.

Granola

Attio

View agent

SALES

Deal intelligence agent

Combine Gong, Attio, and Slack signals to surface deal risks and next-best actions. Updated after every call.

SALES

Outbound prospecting agent

Build targeted prospect lists with Apollo, enrich with firmographic data, and draft personalised outreach. Runs on a schedule.

Apollo

View agent

GTM

HubSpot to Slack updates agent

Watch HubSpot deal stage changes and post structured updates to the right Slack channel. Reps stop checking the CRM all day.

HubSpot

Slack

View agent

ENGINEERING

Engineering standup agent

Aggregate GitHub and GitLab activity, link to Jira, and post a daily standup digest to Slack. No async updates.

OPS

Email-to-calendar scheduling agent

Parse scheduling intent from Gmail threads and create Google Calendar events with the right attendees and timezone.

Gmail

Google Calendar

View agent

Customize the sample

Clone it and own it with connectors you choose

Don't sweat the integration. Point a coding agent at the repo. It clones, swaps in your connectors, and adds new steps for you.

1

Install a coding agent

terminal

Copy Copied

claude "Set up a post-call CRM agent using Granola MCP, HubSpot, Gmail, and Slack via Scalekit"

terminal

Copy Copied

codex "Set up a post-call CRM agent using Granola MCP, HubSpot, Gmail, and Slack via Scalekit"

terminal

Copy Copied

gh copilot suggest "Set up a post-call CRM agent using Granola MCP, HubSpot, Gmail, and Slack via Scalekit"

terminal

Copy Copied

Open Cursor Composer (Cmd+Shift+I) Paste the prompt from the Prompt tab

terminal

Copy Copied

npx skills add scalekit-inc/skills --skill setup-scalekit

2

Give it this prompt

prompt

Copy Copied

Clone github.com/scalekit-inc/python-connect-demos/langchain. Set connection_names = ["granolamcp", "hubspot", "gmail", "slack"]. Build a post-call CRM pipeline: read Granola transcript, extract deal signals with the LLM, update HubSpot deal stage and notes as the rep (Scalekit handles 30-minute token refresh), save a follow-up draft to Gmail, post summary to Slack. Set SCALEKIT_ENV_URL, SCALEKIT_CLIENT_ID, SCALEKIT_CLIENT_SECRET in .env.

Post-call CRM agents that update records on-behalf-of your users

How the agent goes from call transcript to CRM update in five steps

A real working agent you can deploy

Delegated identity. Not service accounts.

Agents that update your CRM, without the auth plumbing

Prebuilt agents you can ship today

Clone it and own it with connectors you choose

Build your own
‍multi-connector agent

Post-call CRM agents that update records on-behalf-of your users

How the agent goes from call transcript to CRM update in five steps

A real working agent you can deploy

Delegated identity. Not service accounts.

Agents that update your CRM, without the auth plumbing

Prebuilt agents you can ship today

Clone it and own it with connectors you choose

Build your own‍multi-connector agent

Build your own
‍multi-connector agent