Six prompts · Copy and paste
Cold email prompts worth asking
Not fifty prompts. Six, each with the reason it is worded the way it is — because the wording is doing the work, and a prompt you cannot reason about is one you cannot fix when the answer comes back wrong.
These are sharper with the Outreach Skill loaded, since the model then answers from a method rather than from whatever it absorbed about cold email on the open internet. They are still useful without it. Installing takes two minutes.
Here is my business and who I sell to. Write me a cold email.
I run [business] in [city]. I sell [service] to [who]. Their usual problem is [problem in their words]. Write me a cold email: 50–125 words, one question as the CTA, no link, nothing invented.The constraints are the prompt. Left to itself a model writes 200 words, opens by describing your company, and closes by asking for a 15-minute call — all three of which cost replies. Naming the length, the single ask and the no-link rule up front gets you a sendable draft on the first pass instead of the fourth.
I sent 400 emails and got 2 replies. What is broken?
I sent [n] emails over [period] and got [n] replies, [n] bounces and [n] opt-outs. Here is the email: [paste]. Here is the follow-up sequence: [paste]. Diagnose it in order of likely impact.The useful part is "in order of likely impact". A 0.5% reply rate has maybe six plausible causes and they are not equally likely — placement before copy, copy before targeting, and the CTA before any of them. Without an ordering instruction you get all six listed as equals and no idea which to fix first.
I need 3 new clients a month at $600 each. Is cold email realistic for me?
I need [n] new clients a month. Average first-year value is [$]. I can spend [hours] a week on outreach. Work the funnel backwards and tell me whether this is realistic, and what volume it implies.This is the question most people should ask first and almost nobody does. Worked backwards from a reply rate, a meeting rate and a close rate, a target either implies a sane weekly volume or it implies four thousand emails a month — and finding that out before you buy a domain is worth more than any subject line. Expect the honest answer to be no more often than yes.
Check this sequence for anything that will get me marked as spam.
Here is my sequence: [paste all touches]. I am sending from [domain], [n] a day, warmed up over [period]. Flag anything that risks placement or breaches CAN-SPAM, and separate the two.Deliverability and legality get conflated constantly, and the fixes are different — a missing postal address is a legal problem, a 400-a-day ramp on a three-week-old domain is a placement problem, and one of those gets you fined while the other just gets you ignored. Asking for them separately is what makes the answer actionable.
What do I need to set up before my first send?
I am about to start cold email from [domain]. Nothing is set up yet. Give me the setup in order, and tell me which parts I can skip at low volume.The "what can I skip" clause is the whole value. Every deliverability guide reads as though you are about to send a hundred thousand emails a month; at thirty a day most of it is theatre and some of it is not. The method knows which is which.
Rewrite this so it does not read like a template.
Here is my email: [paste]. It is going to [n] businesses. Rewrite the opening line so it could only have been sent to this one, and tell me what research that requires.The honest answer is usually "this needs two minutes per prospect and you have twelve hundred of them". A model asked simply to personalise will cheerfully produce fake specificity — praise for a website it cannot see. Asking what research the line requires surfaces the cost instead of hiding it, and for a commodity service the right answer is often to drop personalisation entirely and lead on one idea shared by every prospect.
Why a prompt alone only gets you so far
A prompt shapes one reply. It cannot carry the material the reply needs to be right about — the sender rules as they stand in 2026, the sequencing timings, the arithmetic that decides whether outbound suits a $600 deal. So a good prompt gets a model to behave better while it still knows the same generic things, and the gap shows up exactly where it costs you: it will write you a tidy 90-word email and still suggest putting a calendar link in the CTA.
That is the case for a skill rather than a prompt library: a skill is a reference the model consults, holding roughly 6,500 words it reads as needed, and it applies itself whenever the task involves outreach instead of waiting to be remembered.
It is also worth being clear about what neither fixes. A skill is static knowledge: it cannot tell you who replied yesterday, which of your domains is burnt, or who has already asked not to be contacted. That needs live state, which is a different mechanism — MCP — and a different page.
What running this yourself actually takes
The method is the part people expect to be hard, and it is the part you now have for nothing. The work is everything around it — and almost all of it is a fixed cost you pay whether you send 200 emails a month or 2,000.
| What | To stand it up | Then, forever |
|---|---|---|
| Sending domains | Buy a separate domain, never the one carrying your real mail | Replace any that burn |
| Authentication | SPF, DKIM and DMARC, correct on every sending domain | Re-check after any DNS change |
| Inbox warm-up | Three to four weeks of ramping before a single real send | Re-warm anything that sat idle |
| Lists | Source the companies, find the people, verify the addresses | Rebuilt continuously — lists decay about 2% a month |
| Suppression | One list that holds across email, SMS and calling | Every opt-out, every channel, immediately |
| Placement monitoring | Seed accounts, so you find out before the replies stop | Weekly, and after every copy change |
| Calling or SMS | A2P 10DLC brand and campaign registration, numbers, a dialler | DNC scrubbing, calling windows, spam-flag rotation |
| Writing and sending | The method — which you now have in full | 20–40 hours a month at any real volume |
Below a few thousand sends a month that overhead is the job. And the failure mode is almost never a bad email — it is three good weeks followed by five silent ones, because sequences only produce anything when they actually run, and outbound is the task that slips first when a busy week arrives.
Run it yourself when you have 5–10 hours a week that are genuinely free every week, you are willing to spend the first month warming a domain before a single real send, and somebody will still be triaging replies in month four when it has stopped being novel. Plenty of people meet that. It is completely learnable and nothing is held back.
Hand it off when those hours are worth more spent elsewhere. For most owners of a working service business they are: 20–40 hours a month is $1,500–3,000 of your own time, and the same hours put into delivery or sales calls usually return more than they cost.
Or we run the whole thing
The Outreach Platform sources the leads in your niche, writes and sends the email, SMS, voice and social outreach, owns the deliverability, and forwards replies straight to your inbox. There is no dashboard you have to operate — and because it exposes an MCP server, the same assistant you just installed the skill into can run the account from the conversation you are already in.
| Plan | What runs | Month |
|---|---|---|
| Free | 10 leads and a written pitch for one — nothing sent | $0 |
| Sourcing, writing, sending, deliverability | $29 | |
| + Calling | Adds voice, registration and DNC handling | $59 |
| Full volume | Every channel at full send | $99 |
Set $29 against $1,500–3,000 of your own time and the comparison is the whole argument — run it with your own hourly rate and see where it lands. No contract. And the free tier means you can see the output on your own market before deciding anything.
No credit card for the free tier · Nothing sends until you say so