Check cold emails for spam risk before sending
Catch cold emails likely to land in spam before they go out.
What the agent did
Source-reported example · as described by Federico Donatone
An agency founder has a Claude Code agent score every cold email draft and its sending domain the way a spam filter would, before anything is sent.
The author gives no deliverability or reply-rate numbers.
Limitations
The author notes that a good score on a bad contact list still gets no replies.
Try it in your business
What you'll need
Claude Code on a paid Claude plan and plain-English rules and skill files. The scoring method and any tools used for domain checks are not disclosed.
Starting prompt
The author didn't publish their prompt, so this is our adaptation.
How ours differs from the original setup: Scoring method not published; ours asks for a 10-point score and rewrite on a single draft.
Before I send this cold email, review the draft and our sending domain [domain]. Flag anything likely to trigger spam filters (spammy words, too many links, missing unsubscribe, formatting) and anything that sounds generic. Give me a score out of 10 and a suggested rewrite. Don't send anything.
Replace the [bracketed] parts with your own files and tools.
Get examples like this every Tuesday and Friday
Practical ways to put AI agents to work in your business. We check the original sources, explain the setup and point out the limits. Free.
Unsubscribe any time. Privacy policy
Related use cases
Draft sales-call recap emails from your notes
Have the follow-up email after a sales call drafted from your notes, ready for you to check and send.
Author's example instructions
You'll need: Claude Code on a paid Claude plan, a plain-English rules (skill) file, and call notes saved as files. How the notes reach the agent is not disclosed.
1. Read the call notes first. Use only names, numbers and dates that appear in them. 2. Write in my voice: short sentences, first person, no buzzwords, no exclamation marks. 3. Open with the one problem the prospect cared about most. 4. List the next steps, each with an owner and a date. 5. Never send the email. Save a draft and tell me where it is. 6. If the notes are missing or unclear, stop and ask me instead of guessing.
Get a morning list of deals that need you
Start the day knowing which open deals need your attention, without checking the CRM yourself.
Author's related starter template
You'll need: Claude Code on a paid Claude plan, plain-English rules and skill files, read access to your CRM, email and WhatsApp, and a schedule. Which CRM, how it is connected and how priority is decided are not disclosed.
| Section | Example line | | --- | --- | | Job | Every Monday at 8:00, write a one-page pipeline summary for the founder. | | Inputs | Open deals in the CRM, read only. Last week's call notes in /notes. | | Voice | Short sentences, first person plural, numbers before adjectives. | | Never | Send email, change a deal stage or delete a file. | | Done means | Every open deal is listed with its next step, owner and date. | | Mistakes log | YYYY-MM-DD: listed a closed deal as open. Check the stage field first. |
Build a cold email campaign
Go from a target list to cold email campaigns set up in your sending tool from a short request.
Our suggested starting prompt — not the author's original · untested
You'll need: Claude Code connected to AI Ark, Clay, an email-finding tool (named only as "LM") and Bison. The prompts, how the tools were connected and any review steps are not disclosed.
Help me set up a cold email campaign for [offer] aimed at [ideal customer]. Pull a list of [N] matching companies from [lead tool], check each one against my criteria [list] and drop the ones that don't fit, find a contact email for the right person, and draft a 3-step sequence. Load everything into [sending tool] as a paused draft campaign. Don't start sending; show me 5 sample emails first.

