# An email sequence that only says what you can prove

> An email sequence that only says what you can prove. Drafts only, never sends.

Loop: `sequence-from-brain`
Category: Demand
Autonomy: assisted
License: MIT
Source: https://vibemarketers.ai/loops/sequence-from-brain

Drafts a nurture sequence from your positioning and buyer questions, then checks every claim in it against the claims file. Drafts only, no send path.

## What you need

- none required

## Goal prompt

Give this to your agent. Replace anything in <angle brackets> with your own.

```
Read the positioning, ICP, buyer questions and claims file for <company>.
Draft a sequence of emails, each with one job and one call to action. Every
factual claim in every email must already exist in the claims file. Write
the drafts to a drafts directory. Never send, never connect to an email
platform, never schedule anything.
```

## Verifier prompt

Run this as a second, separate agent. It grades the first one's work and must be
able to fail. A checker that never fails is not a checker.

```
Read every draft against the claims file. FAIL if any email asserts a fact
that is not in the claims file, if any email has more than one call to
action, if any send or schedule path exists anywhere in the loop, or if a
draft left the drafts directory. Otherwise PASS.
```

## Stopping condition

Every email in the sequence exists as a draft with every claim verified against the claims file, no send path exists, and the verifier marks PASS.

## How to run it

1. Save this file into your project, for example `.claude/loops/sequence-from-brain.md`.
2. Open your agent in that project (Claude Code, Cursor, or Codex).
3. Paste the goal prompt, with your own values in place of the angle brackets.
4. When it stops, start a fresh session and paste the verifier prompt.
5. Read the verdict before you accept any change.

## What happened when we ran it

Last run 2026-09-25 on vibemarketers.ai.

- Baseline: 17 provable claims in the claims file and no sequence written from them. Any email written by hand would have had no automated check that its claims were among them.
- Result: 5 emails drafted using 4 claims, with 0 claims that were not already in the claims file. 0 sends: the loop has no send path. Two of the five emails assert nothing measurable and say so.
- What this does not prove: Verified claims do not make a good email. This shows the sequence says nothing unprovable, not that anyone would read it or reply to it. No email was sent, so there is no open rate, no reply rate and no evidence of effect.
- Cost: 0 human touches, 7 minutes, $0.00
- Raw log: https://github.com/genwfurukawa/vibemarketers-v2/blob/main/lab/runs/2026-09-25-sequence-from-brain.json

## Inside a company

- Problem it solves: Nurture sequences drift into claims nobody checked, because the person writing them is not the person who knows what is provable.
- Source of truth it reads: The positioning, ICP, buyer questions and claims and proof files.
- What it may do: Read the Brain, write drafts to a drafts directory. Never send, never schedule, never authenticate to an email platform.
- Who approves: Marketing approves every email before it is loaded anywhere.
- Who owns it: Demand generation lead
- How you measure it: Share of drafted emails that ship without a factual edit.

MIT licensed. Copy it, change it, run it. No attribution required.