Welcome to 314 members who joined this week 👋

Hey there, I’m Joshua, and welcome to the NCF Weekly newsletter where we share actionable guidance on building a business as a non-technical founder -- using no-code tools and AI.

Check out our other newsletters: Tiny Empires and Seedstrapped

Hyperagent is a browser-based platform from the Airtable team that lets you build AI agents that connect to your existing tools and run on a schedule or on demand. This guide covers how to set one up, what the key concepts mean, and what the agents can actually do.

Hyperagent has kindly given the first 1,000 NCF readers get 1,000 free credits. You can claim these here.

System prompt. The agent's standing instructions. It loads every time the agent runs and defines what the agent does, how it approaches tasks, and what it should ask you when it needs more context. You don't write this from scratch. Describe the job in plain language and Hyperagent drafts the prompt for you.

Skills. Modular instructions for specific tasks. Unlike the system prompt, skills only load when the agent decides they're relevant. A chief of staff agent might have one skill for daily briefings, another for meeting prep. Skills are documented in plain language and can include scripts when needed. You create them by starting a thread with your agent and describing what you want it to know how to do.

Invocations. What triggers the agent to run. Options are: a set schedule (e.g., every morning at 7am, or every 5 minutes to check an inbox), a Slack mention, or a direct thread. Most agents use a combination.

Integrations. What the agent has access to. Gmail, Google Calendar, Slack, Airtable, HubSpot, Salesforce, and hundreds more. Connect whatever data sources the agent needs to do its job.

When you log in, you'll see six pre-built starter agents: chief of staff, data analyst, developer, investment analyst, recruiter, and sales prospector. Each comes with a system prompt already written and is ready to connect to your tools.

If none of those fit, go to New Thread and describe the job you need done in a paragraph. Hyperagent will research context about your company, generate a system prompt, suggest skills, and recommend integrations before creating the agent. You can edit any of it before saving.

The chief of staff is the most useful starting point for most founders. Connect it to Gmail, Google Calendar, and Slack, set a scheduled invocation for the morning, and it will deliver a daily briefing to a Slack channel: what's in your inbox, what's on your calendar, any relevant news about competitors.

A few concrete examples from a recent Hyperagent demo:

VP marketing agent. Connected to an Airtable base with campaign data, lead records, and pipeline stages. When asked to run a daily pipeline review, it pulled data across multiple tables, calculated conversion rates by channel, and produced a report with specific recommendations. One finding: conferences were consuming 35% of spend but returning a lower ROI than other channels, so it flagged that for reallocation. The analysis came from raw data only, no formulas pre-built, no template set up in advance.

BDR inbox agent. Set up to monitor a Gmail inbox and handle inbound leads by category. Hot leads got an immediate reply with a Calendly link. Neutral replies got flagged to a Slack channel for a human to handle. Unsubscribe requests got marked as do-not-contact. The agent had one skill loaded with examples of how the company's sales team typically writes, which kept the tone consistent without it sounding templated. Each email it processed triggered a Slack update summarizing what it did and why.

Both were built from a single paragraph description with no prompt engineering involved.

Agents can update themselves based on feedback in three ways.

Skills. If you tell an agent to always do something differently, it can update the relevant skill so that change sticks. You don't need to edit the skill manually.

Memories. Preferences you express in conversation get saved automatically if you enable autosave, or you can review and approve them manually. These memories apply to every future interaction with that agent.

Evaluations. You can ask an agent to generate a scoring rubric based on its own system prompt, then run its outputs against that rubric. This is useful for agents running the same workflow repeatedly where you want to track and improve quality over time.

All three are visible in the agent settings, where you can inspect and edit what the agent has saved.

Pick one recurring workflow that requires judgment but follows a loose pattern. Responding to a certain type of email, pulling a weekly report, prepping for a recurring meeting. Write a paragraph describing what you'd want a new hire to do, paste it into a new thread, and let Hyperagent build from there.

Get started with Hyperagent with 1000 free credits for the first 1000 NCF members here.

Get full access to all our resources with Pro Membership for a one-time payment of $149.

Sneak peek at what’s included 👇

💰$50k in Perks - Access Exclusive offers from the top no-code tools who have partnered with us. Including $15k waived Stripe fees, $500 Bubble credit + $1000 Coda credit View all

💻 No-Code Operating System - Advanced Notion template that replaces all your productivity tools. Everything you need to launch and manage your business from one page. Learn more

🎓 Course: Tiny Empires Method - Learn how to build a 6-figure business that works around your life and not the other way around. Learn proven frameworks to stop wasting time and start making money. Learn more

🎓 Course: Sales for Introverts (and people who don’t like selling) - Learn how to sell in a way that fits your personality and delivers consistent, reliable revenue for your busines. Learn more

🛠️ 75+ Curated No-Code Courses and Resources - We’ve curated 2000 no-code videos into 75+ easy to navigate courses. Save yourself hours of watching Youtube videos that don’t move your knowledge forward. View all