Workflows · facts checked 2026-09-13

7 Town agent workflows for founders

A curated list of 7 Town workflows for founders, each with the inputs it needs, a starting prompt, the expected output and one honest boundary.

Seven founder workflows worth handing to Town, each with inputs, a starting prompt, the output to expect and the boundary. Matching $ready-made AI agents are linked where they exist.

1. Investor target queue

When to use it: When the raise starts and the list is a blank sheet.

Inputs: Your fundraising thesis: stage, sector, geography, check size.

Starting prompt:

You are a fundraising researcher. Build a target investor list for this startup. Give me [20] investors who actually fund this stage and sector - angels and small funds included, not just famous names. For each: name, fund, what stage they write checks at, 1-2 portfolio companies close to what I do, why they're a fit in one line, and where to find them (LinkedIn or fund site). Only include investors you have evidence for - if you're not sure they invest at this stage, mark it "verify". Rank by fit.

My startup: [what you do, stage, sector, geography]
Raise: [amount, round type]
Investors I already know about: [paste any]

Expected output: A ranked, evidence-backed list with verify flags.

Boundary: It can be wrong or stale - verify each investor before outreach.

Ready-made: the Investor Outreach Agent runs this as a standing workflow with gates, free download.

2. Weekly KPI update

When to use it: Every Friday, before the weekend eats it.

Inputs: Your approved metrics sheet, last week’s update for format.

Starting prompt:

You are a founder writing the weekly update to [investors / advisors / team]. Structure: 3 numbers first (the metrics that matter this week), then one win, one problem and what I'm doing about it, and one ask (intro, hire, feedback). Under 150 words total. No greeting, no "hope you're well", no spin - investors trust founders who say the bad thing plainly. If a number went down, state it and the reason in the same sentence.

This week's numbers: [..]
The win: [..]
The problem: [..]
The ask: [..]

Expected output: A paste-ready update in founder voice.

Boundary: Your reputation - keep it honest.

Ready-made: the KPI Update Agent runs this as a standing workflow with gates, free download.

3. Customer discovery themes

When to use it: After every 5-10 interviews, before the insights evaporate.

Inputs: Anonymized interview notes.

Starting prompt:

You are a customer researcher. Read these interview notes and pull out: the top 3 problems people described (with a quote for each), how many people mentioned each, the exact words they used (I'll reuse them in marketing), anything that surprised you, and what people currently do instead. Then: 3 things this suggests we should build or change, each tied to the evidence. Keep quotes verbatim. If the sample is too small to conclude something, say so.

The notes: [paste notes from all calls, separated]

Expected output: Patterns backed by real quotes, not your vague memory.

Boundary: It reads these notes; it can't hear tone.

Ready-made: the Customer Discovery Agent runs this as a standing workflow with gates, free download.

4. Partnership target queue

When to use it: When partnerships are the growth channel and the list is stale.

Inputs: Your partnership thesis: who, why you, what you offer.

Starting prompt:

Write a partnership email to this company. Rules: first line shows I know what THEY do (use what I paste), then the specific overlap - what their customers want that we provide. Make the ask tiny: a 20-minute call or a pilot with [specific small scope]. One line on what's in it for them, quantified if I give you numbers. Under 90 words, no "synergy", no "reaching out", no flattery.

Who they are: [paste their About or site text]
What we do: [one line]
The overlap I see: [..]
The small first step I'm offering: [..]

Expected output: A short email a busy exec reads to the end.

Boundary: Only offer what you can actually deliver.

Ready-made: the Partnership Outreach Agent runs this as a standing workflow with gates, free download.

5. Board meeting prep

When to use it: The week before each board meeting.

Inputs: Your metrics, last board deck, the decisions you need.

Starting prompt:

You are a founder prepping the board meeting. From my numbers and notes, draft: the 1-paragraph CEO summary (state of the company, no spin), 5 slides with 3 bullets each (metrics, wins, misses, plan, asks), the 2 hard questions the board will definitely ask with honest draft answers, and a pre-read email under 100 words. If a number is missing, flag it - boards smell gaps.

Numbers: [paste]
Big items this quarter: [..]
The ask: [..]

Expected output: A deck skeleton, hard questions, and the pre-read.

Boundary: You face the board; the answers must be true.

Ready-made: no agent for this one yet.

6. Hiring scorecard builder

When to use it: When you are about to interview for a role you have never hired.

Inputs: The role, what success looks like at 90 days, your non-negotiables.

Starting prompt:

You are a hiring lead. Turn this role into an interview scorecard: the 5 things that actually predict success here (not generic "culture fit"), each with what a strong answer looks like, what a weak one looks like, and a 1-5 anchor scale. Add 2 knock-out criteria (auto-no) and 2 "nice but trainable" items so interviewers don't confuse them. One page, printable.

The role: [paste JD or 3 lines]
What great looks like in the first 90 days: [..]

Expected output: A printable scorecard with anchors and knock-outs.

Boundary: You calibrate it against the first few interviews.

Ready-made: no agent for this one yet.

7. Runway scenario check

When to use it: Monthly, or before any big spend decision.

Inputs: Current cash, monthly burn, committed spend, revenue rows.

Starting prompt:

You are a founder's finance analyst. From these numbers, model 3 runway scenarios: current burn, cut 20%, and "if we close [deal or raise] by [date]". For each: months of runway, the month cash hits zero, and the 2 decisions that move the date most. Then the uncomfortable paragraph: what I should do this month given the worst case. Plain words, no false precision.

Cash today: [$]
Monthly burn: [$]
Revenue and growth: [..]
Possible events: [raise, big deal, hire plans]

Expected output: 3 scenarios and the honest worst-case paragraph.

Boundary: It models what you give it; the decisions are yours.

Ready-made: no agent for this one yet.

Quick comparison

WorkflowEffortCadenceRiskReady-made agent
1. Investor target queueOne paste per runPer eventDraft-first, review before useInvestor Outreach Agent
2. Weekly KPI updateOne paste per runWeeklyDraft-first, review before useKPI Update Agent
3. Customer discovery themesOne paste per runPer eventDraft-first, review before useCustomer Discovery Agent
4. Partnership target queueOne paste per runPer eventDraft-first, review before usePartnership Outreach Agent
5. Board meeting prepOne paste per runWeeklyDraft-first, review before usenone yet
6. Hiring scorecard builderOne paste per runPer eventDraft-first, review before usenone yet
7. Runway scenario checkOne paste per runMonthlyDraft-first, review before usenone yet

Free download

Get the Investor tracker template (CSV) - the companion file for these workflows, with sample rows filled in.

Platform notes

Town is a third-party product (https://www.town.com). We are not affiliated with or endorsed by them. Capabilities and names verified 2026-09-13; re-check before relying on specifics. The prompts above are platform-agnostic - they work in any capable assistant.

Want the job done instead of prompted? Browse the founders agent menu or the AI agents for founders guide.

Questions people ask

Are these Town workflows free?

Yes - every item includes a starting prompt you can copy free. Matching free standing agents are linked where they exist.

Are you affiliated with Town?

No. AI Agent Skills is an independent store; Town is named only because these workflows run on it. Facts checked 13 September 2026.

Last updated 13 September 2026 · Reviewed by the AI Agent Skills team.