cold-email-agency
Skill0 starsChanged 22 days agoMITView on GitHub
Write B2B cold emails and 4-step follow-up sequences that sell AI/automation services to US SMBs. Use when the user asks for cold outreach, a prospecting sequence, or reply-rate fixes.
In this repository4 other files it gives its agents.
- client-workspaceSkill
- scope-guardSkill
- offer-builderSkill
- quick-quoteSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
No AI buzzwords above the fold for non-tech buyers: say "answers every call in 2 rings", not "AI voice agent".
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Cold Email for AI Agencies
Goal: get a reply, not a sale. One email = one idea = one ask.
Inputs (ask if missing)
- ICP: industry, company size, role (e.g. "owner of a 5-20 person HVAC company in Texas").
- The ONE pain you solve in their words (missed calls, slow quotes, no follow-up, manual data entry).
- Proof: a number, a case, or a demo (if none, use a free audit/teardown as proof).
- The ask: low-friction (a 2-min Loom, "worth a look?", a free audit), never "book a 30-min call" on email 1.
Rules
- 50-90 words. Subject 2-4 words, lowercase, no clickbait ("missed calls", "quick idea for [company]").
- First line is about THEM (a trigger: hiring post, reviews, site speed, ad they run). Never "I hope this finds you well".
- No AI buzzwords above the fold for non-tech buyers: say "answers every call in 2 rings", not "AI voice agent".
- One CTA, phrased as a yes/no question.
- Plain text, no images, no tracking pixels on step 1 (deliverability).
- Personalization variables:
[first_name],[company],[trigger],[city].
Sequence (default)
| Step | Day | Angle |
| 1 | 0 | Trigger + pain + soft ask |
| 2 | 3 | Proof: mini case or number |
| 3 | 7 | Give value: 1 specific fix they can do without you |
| 4 | 12 | Breakup: "should I close the loop?" |
Output format
For each step: subject, body, and 2 A/B variants of the first line. Then a checklist: SPF/DKIM/DMARC, warmed inbox, <40 sends/inbox/day, separate sending domain.
Example (step 1)
Subject: missed calls
Hi [first_name] — saw [company] has 140+ Google reviews but the phone went to voicemail when I called at 6pm. For HVAC that's usually a $400+ job going to the next listing. We set up a system that answers and books those calls automatically. Worth a 2-min video showing how it'd work for you?
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.