growthenginenowoslawski/coldoutboundskills

# Auto Research (Public)

Automated end-to-end campaign launcher. Feed it one target company domain, get back a live Smartlead campaign with per-lead personalization — in about 20 minutes.

This is the beginner-friendly version of the GEX internal `auto-research-v2`. All state lives in local JSON files; no Supabase, no Trigger.dev.

## What you get

- **Input:** one target company domain + your `client-profile.yaml`
- **Output:** a running Smartlead campaign with:
  - 200-1,000 leads (depending on targeting tightness)
  - Per-lead personalization: 9 custom variables (situation, value, CTA × 3 variants)
  - A/B/C subject + body variants tested in parallel
  - Campaign assigned to your available inboxes

# Blitz List Builder

Prospeo is title-first: "all VPs of Marketing in B2B SaaS." Blitz is domain-first: "for each of these 500 restaurant domains, find the owner."

If you already know the target companies (from Google Maps, a CSV of competitors, or a manually-curated list), Blitz finds the people for you.

## When to use Blitz vs Prospeo

| Your situation | Use |
|---|---|
| "I have 5,000 domains, find the owners" | Blitz |
| "Find all CFOs at US SaaS companies 100-500 employees" | Prospeo |
| "I need to supplement Prospeo results with additional contacts at the same companies" | Blitz |
| "I want the widest net possible in a vertical" | Prospeo |

# Campaign Copywriting Skill

You are a cold email copywriter. Your job is to create high-converting cold email campaigns through a **stepwise confirmation process** that ensures alignment before outputting final copy.

## How This Skill Works

This skill operates like Typeform—you confirm key decisions at each step before proceeding. This prevents overwhelming output and increases the likelihood that final copy will be accepted.

**The 4-Step Flow:**
1. **Confirm Campaign Direction** - Research, summarize, get approval on overall approach
2. **Confirm Subject Line + First Line Strategy** - Present options, get approval
3. **Confirm Body Structure** - Value prop, case study, AI variables, CTA style
4. **Output Final Copy** - All variants, follow-ups, ready to paste

# Campaign Strategy Skill

You are a cold outbound campaign strategist. Your job is to generate **at least 15-20 campaign ideas** (more if you have strong ideas—never cut good ideas to hit an arbitrary limit) that range from broad targeting to highly niche targeting, each with a clear AI personalization strategy and value proposition.

## Core Philosophy

Every campaign has two levers:
1. **The List** - Who we're reaching out to (broad → niche)
2. **The Message** - What value proposition we're leading with

The deeper and more focused the list, the more the messaging should relate to that specific list. Broad lists require AI-generated personalization to feel relevant. Niche lists can reference the filtering criteria directly.

### Value Proposition Categories

# Cold Email Kickoff

The single entry point for a fresh user. Replaces "stare at 29 skills and guess which to run first."

## When to use

- First-time cold email operator on this repo
- New client/business being onboarded (fresh `profiles/<slug>/` directory)
- Experienced operator starting a new campaign for a new business
- Anyone who asks "where do I start?"

## What this skill produces

- A populated `profiles/<business-slug>/client-profile.yaml` (from `/icp-onboarding`)

# Cold Email Starter Kit

A complete system for running cold email. Follow it top-to-bottom and you will have purchased domains, created warmed inboxes, built a lead list, written a campaign, and sent your first emails. Designed for beginners — no prior cold email experience required.

---

## What you'll build

By the end of the flow you'll have:
- 10-20 throwaway sending domains purchased on Dynadot
- 20-40 warmed sending inboxes on Zapmail
- A 2,000-person lead list from Prospeo
- A 4-step email sequence with variants and personalization
- A live campaign in Smartlead or Instantly sending daily

# Cold Email Weekly Rhythm

Having 29 cold email skills doesn't help if you don't run them on a schedule. This is that schedule.

**This is a pure playbook.** No scripts, no automatic reminders. You put the rhythm on your own calendar (Step 1 below) and run the prescribed skill at the prescribed time. That's the entire system.

---

## Step 1 (required before using this skill): Put the rhythm on your calendar

Open Google Calendar / Outlook / Apple Reminders / whatever you actually look at every day. Create these as recurring events. Copy the titles and cadences exactly.

| Event title | Cadence |
|---|---|

# Competitor Engagers

Finds people who are actively engaging with your competitors on LinkedIn — the highest-intent prospects for cold outreach.

## Required step: Qualify with /icp-prompt-builder (do not skip)

After collecting the engager CSV, sample ~50 rows and run `/icp-prompt-builder` before scaling further. LinkedIn engagement alone doesn't guarantee ICP fit — someone who reacted to a competitor's post might be a peer, a candidate, a student, or an actual buyer. Qualifying filters separates them.

**Why required:** unfiltered engager lists typically have 30-50% non-ICP rows (fans, peers, recruiters, students). Running `/icp-prompt-builder` on a 50-row sample, tuning the qualification prompt, then applying it to the full list cuts wasted enrichment + send costs. Takes 10-15 min.

## Setup (First Time Only)

Before running, ensure these environment variables are set in `~/.env`:

# Deliverability Incident Response

Things break. When they do, you need a playbook — not panic. This skill is the triage decision tree.

## The five incident types

| Symptom | Most likely cause | First action | Fix time |
|---|---|---|---|
| Reply rate dropped sharply | Deliverability — emails in spam | Run spam placement test | 1-14 days |
| Bounce rate spiked >3% | Bad list OR domain reputation | Check bounce types | 1-3 days |
| Domain blacklisted | Shared IP bad actor OR your domain got flagged | Check blacklists, rotate if needed | 7-30 days |
| Inbox blocked in warmup | Warmup network flagged sending patterns | Pause, investigate, maybe replace | 1-7 days |
| Gmail marking as promotional | Content triggers (links, images, HTML) | Simplify content | 1-3 days |

# Deliverability Test — Inbox Type Comparison

Compare reply rates, bounce rates, and positive reply rates broken down by inbox type (SMTP, Gmail / G Suite, Outlook / Office365) across your Smartlead account. Use it to answer "are my Gmail inboxes performing better than my SMTP ones?" or "why is one inbox type bouncing more than the others?"

## Why this exists

Different inbox types have different deliverability characteristics:
- **Gmail / G Suite** — highest trust, fewest bounces, often best reply rates
- **Office365** — good reputation, but stricter spam filtering
- **SMTP** (Zapmail, Maildoso, Mailforge, custom) — varies wildly by provider, often has higher bounces in the first 30 days of domain life

If your campaigns mix inbox types, this skill shows which type is earning its keep.

## What you need

# Disco-Like

Lookalike company discovery. Give it 3-10 seed domains you know are a good fit; it returns hundreds of similar companies by domain, industry, and business characteristics. Useful for expanding from a small known-good list to a much bigger TAM without manual research.

## When to use

- You have 3-10 customer domains you love, want "more like these"
- You want to expand a small client list into a full TAM
- You have an ICP description but don't want to manually build Prospeo filters
- Competitive / adjacent-market expansion

## When NOT to use

- You need PEOPLE, not companies (use Prospeo or Blitz after this)

# Email Deliverability Audit

**If your positive reply rate is dropping and you don't know why, start here.** Most of the time the problem is deliverability — your emails aren't reaching inboxes. This skill tells you what's broken.

## What it checks

| Layer | What | How |
|---|---|---|
| DNS auth | SPF, DKIM, DMARC present on each sending domain | `dig` commands |
| Inbox health | Warmup status, reputation, blocks, connection failures | Smartlead email-accounts API |
| Volume health | Daily sent trending, capacity utilization | Smartlead analytics |
| Send + reply rate per inbox | Sent count, reply count, reply rate % over lookback period | Smartlead campaign analytics |
| Bounce rate | Per-inbox and per-domain bounce rate over last 30 days | Smartlead campaign analytics |
| Spam placement | Real inbox-vs-spam test via Smartlead Smart Delivery | optional |

# Experiment Design

If you change your list, your copy, and your offer at the same time, you learn nothing. This skill forces you to isolate one variable per experiment so you actually learn what's working.

## Why this exists

Most cold email operators run "throw-everything" experiments. Campaign 1 gets a new list, new copy, and a new offer. It works better. They declare victory. But they can't tell you WHY — was it the list? The copy? The offer?

Then campaign 2 changes all three again. Regression. Nobody knows why.

This skill is the antidote: plan each experiment around ONE variable, keep everything else constant, and confidence-weight the results.

## The three experiment types

# Google Maps List Builder

A self-contained tool for scraping business listings from Google Maps. Give it a search query (e.g., "pizza restaurant") and a location (zip code, city, or coordinates), and it returns structured data for every matching business — written to CSV.

## How this fits in the cold email flow

Google Maps gives you COMPANIES (name, domain, phone, address, ratings). It does NOT give you PEOPLE. To run cold email:

1. Run this skill → CSV of businesses with `company_domain`
2. **Run `/icp-prompt-builder` on a sample of 50** — tune a qualification prompt to filter out bad fits before paying for downstream enrichment
3. Run `/blitz-list-builder` with the filtered CSV → adds owners/managers to each business
4. Run `/list-builder` (Phase 5, emails) → fills in missing emails
5. Run `/cold-email-starter-kit`'s `smartlead-add-leads.ts` → upload to Smartlead

# ICP Onboarding

Conversational intake. The user arrives with a business; this skill produces a `client-profile.yaml` every other skill in the repo consumes. Without it, downstream skills guess at targeting and write generic copy.

## Why this exists

Most cold email fails because the sender didn't define the ICP tightly enough. "B2B SaaS founders" is not an ICP. "VP of RevOps at 50-500 person B2B SaaS companies in the US that raised Series B in the last 12 months" is an ICP — you can put it into Prospeo and get a list.

This skill forces that precision up front. It also separates **hard filters** (must match) from **soft preferences** (nice-to-have), because the #1 mistake beginners make is treating every ICP criterion as required, ending up with a list of 200 leads instead of 5,000.

## Inputs

Either:
- **Website URL** of the user's business (skill will scrape + infer a lot)

# ICP Prompt Builder

Before you pay to pull 5,000 companies, tune a qualification prompt on 10-50 of them. This skill walks you through the iterative loop.

## Why this exists

List-builder skills (DiscoLike, Blitz, Prospeo, Google Maps) return COMPANIES, but they don't know whether those companies match your ICP. If your list-builder returns 5,000 companies and 80% are wrong fits, you'll waste money enriching them for emails that go nowhere.

The fix: build an AI qualification prompt BEFORE scaling. Pull 10 companies, have the prompt score them, compare to your judgment, refine, repeat. Once the prompt agrees with you 2 rounds in a row with zero corrections, lock it in and apply it at scale.

## Always uses Task sub-agents (no API key)

This skill runs entirely inside Claude Code via the Task tool. No Anthropic SDK calls, no OpenAI calls — Claude Code does the scoring itself. This is intentional:

# Inbox Lifecycle Manager

**Diagnosis tells you a domain is sick. This skill decides whether it dies.**

Every cold email fleet leaks money in two directions at once: you keep paying for burned domains
that will never reply again, and you run under your send goal because nobody promoted the warmed
reserves sitting idle. This is the weekly loop that fixes both, with an approval gate in the
middle so nothing gets cancelled on a hunch.

**Read-only by default.** The plan step writes nothing, anywhere. Only the apply step, run after
an explicit human "yes", changes a tag, a campaign, or a provider subscription.

## When to use

# Lead Magnet Brainstorm

Cold emails with a concrete free offer outperform "book a call" asks by 3-10x. This skill helps the user find an offer they can actually deliver.

## Why this exists

Most cold emails fail the "why should I reply?" test. The answer "because you might need our product" is not compelling. The answer "because we'll send you a free audit that'll save you money whether or not you buy from us" is.

A good lead magnet is:
1. **Cheap to deliver at scale** (ideally automated)
2. **Genuinely valuable to the prospect** (they'd pay for it, or it saves them obvious time/money)
3. **Demonstrates your competence** (so buying becomes the natural next step)

## When to use

# List Builder — the meta list-building process

## ⚡ OPERATE VIA THE RUNBOOK

The manual phase-by-phase flow below still documents the concepts, but the OPERATIONAL
path is now three one-command scripts — see **RUNBOOK.md in this skill directory** for
the decision table and failure playbook. In short:

```bash
npx tsx scripts/make-judge.ts --spec=judge-spec.json --out=prompt.txt   # judge (mandatory blocks)
npx tsx scripts/run-lane.ts  --config=lane.json                         # full lane, resumable
npx tsx scripts/snowball.ts  --config=lane.json                         # sweep until dry
npx tsx scripts/contacts.ts  --config=lane.json                         # contacts + emails after approval
```

# List Expander — seed companies → lookalikes → mined filters → qualified TAM

## The problem this solves

Niche lists come out tiny because we search databases with narrow explicit keywords, but real
good-fit companies (e.g. Atlantic Medical Group, Hackensack Meridian) often don't carry that
keyword in their database record. Instead of guessing keywords top-down, this skill works
bottom-up: take companies you KNOW fit, discover how the database actually tags them, expand via
lookalike engines, and only then derive the wide-net filters — with measured precision per filter
— before pulling and AI-qualifying at scale.

## Pipeline (5 phases)

All scripts live in `scripts/`, run with `npx tsx`, and are dependency-free (node ≥ 18 native

# List Quality Scorecard

A CSV of 5,000 leads is not the same as a good list of 5,000 leads. This skill grades your list across 8 dimensions BEFORE you send, catching preventable waste.

## Why this exists

The three campaign failure modes cold email runners hit most:

1. **Bad list** — the copy doesn't matter when you're emailing the wrong people
2. **Unverified emails** — bounces burn domain reputation
3. **ICP drift** — you think you're targeting VPs, but the list is mostly Managers

Each of these is catchable before you send, in 5 minutes, for free.

# Perfect company list

Database filters fail in two directions at once.

- **They miss real fits.** Industry labels are self-reported, or guessed: many LinkedIn company
  pages carry the banner "This listing was automatically created by LinkedIn", which means nobody
  at the company picked the label. We found an FDIC bank listed as *Construction*, an SEC-registered
  wealth manager listed as *Semiconductor Manufacturing*, and a New Jersey school district listed
  as *Retail Luxury Goods and Jewelry*.
- **They pull in junk.** A plain `Banking` + `United States` filter returned 16,895 companies. About
  10,500 of them were not banks: 3,679 mortgage companies and lenders, 2,414 credit unions, 2,210
  dead sites or unrelated businesses (a Toyota dealer, a pawn and gun shop, a youth soccer club),
  839 vendors that sell to banks, and the Federal Reserve Bank of St. Louis.

# Personalization Sub-Agent Pattern

Cold email personalization at scale requires per-lead generation. Claude Code's Task tool lets you fan out to many sub-agents in parallel, each personalizing a slice of the lead list. This skill defines the reusable approval-loop pattern.

## Always Task tool — never an API key

This skill runs entirely inside Claude Code via the Task tool. No Anthropic SDK calls, no OpenAI calls. This is intentional:

- **No extra API spend.** Uses your Claude Code plan.
- **No key management.** Works out of the box.
- **Parallel by design.** Claude Code spawns multiple Task sub-agents in one message, letting 100 leads finish in the time it takes to personalize 10.

At very large scale (1,000+ leads), the tuned prompt can optionally be shipped to the Anthropic API for throughput. But TUNING and normal campaign runs (under 500 leads) always go through the Task tool.

# Clay Playbooks

Nineteen playbooks. Each one answers a single question:

> A campaign needs signal X on every row. What is the cheapest reliable way to get X, and what exact prompt turns X into copy-ready text?

One playbook produces one field. `new_in_role_line`. `funding_line`. `pricing_page_line`. That
field slots into one sentence of email 1. That is the whole product.

## The three run modes

Every playbook directory ships the same three files. They are three delivery mechanisms for the
same recipe, not three different recipes.

# Playbook: Meta Ad Library

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the angle depends on the prospect spending money on paid social right now — creative
agencies, UGC shops, CRO and landing-page services, media buyers, attribution and analytics tools,
or anyone selling to ecommerce brands that already advertise.

**Do not use when:** you only need to know a company exists or sells online (`playbook-tech-on-website`),
or you want LinkedIn organic activity (`playbook-linkedin-engagement`). Facebook and Instagram only —
not Google Ads, not TikTok, not LinkedIn ads.

**One-line output:** `ad_library_line = "you are using ai to run outbound campaigns with one command"`

# Playbook: AI Specificity Line

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the offer is good but the email reads like it was sent to 5,000 people, and you need one
sentence that names what the client would actually do for **this** company.

**Do not use when:** you want a first line about a person or an event. **Those playbooks personalize
the opener. This one personalizes the offer** — which is a different and usually more durable move.

**One-line output:** `specificity_line = "know the landed cost of every pool float before you price it"`

## 1. Trigger and scope

# Playbook: Case-Study Page Name-Drop

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the campaign wants a first line that proves you looked at the prospect's site, and the
prospect publishes customer stories — B2B software, agencies, consultancies, MSPs, staffing,
manufacturers with named clients.

**Do not use when:** you want a tech detail (`playbook-tech-on-website`), their pricing model
(`playbook-pricing-page`), or an ad they are running (`playbook-ad-library`).

**One-line output:** `case_study_line = "your work with Intercom on outbound pipeline"`, rendering as
`Saw your work with Intercom on outbound pipeline.`

# Playbook: Company Name Cleaning

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

## 0. Read this first: the AI prompt is a toggle, not the default

Clay ships a free, deterministic `Normalize Company Name` action (`titleCase: true`). **That is
the default and it stays the default.** Do not replace it reflexively, and do not make AI the
default.

Everything below — the locked prompt, the benchmark, the placeholder guard, the AI column — is the
**toggle**: what you switch on for one client, when the copy leans hard on the company name **and**
the raw strings are messy in ways a normalizer structurally cannot fix. Those are taglines after a
pipe, dba entities, parenthetical descriptors, appended city names, second-language duplicates, and

# Playbook: 3-Bullet Creative Ideas

> Best practice, not law. Override when the campaign calls for it; note the practice once and proceed.

**Use when** the premise is "I looked at your business and had three ideas."

**Not when** one personalized sentence tops an otherwise fixed email (`playbook-ai-specificity`), or
the hook is an event.

**Output:** `creative_idea_1 = "a production scheduler that plans injection molding runs against cleanroom capacity"`

## 1. Slot discipline is the whole idea

**The operator names what bullets 1, 2 and 3 are about. Those slots stay fixed for every lead, and

# Playbook: First Name Cleaning

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** any campaign whose email opens with the prospect's first name, which is every campaign
anyone runs. Run it next to the company-name clean, **before any other custom variable**, because
variables often interpolate the cleaned name.

**Do not use when:** you need the company name cleaned — that is
`playbook-company-name-cleaning`, its twin, built to the same shape on purpose. Also not for
finding a missing name. This playbook never looks anything up; a blank stays blank.

**One-line output:** `first_name_clean = "Ruba"` from the raw string `"Dr Ruba"`, so the email opens
`Hi Ruba,` instead of `Hi Dr Ruba,`.

# Playbook: Fundraising

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the angle depends on the prospect having money to spend or a mandate to grow — either
a TAM filtered to funded companies, or a first line that names their round.

**Do not use when:** you want headcount growth or open roles as the signal (`playbook-hiring-surge`).
If you want the round as a pure list filter and no copy at all, you only need lane A in §3.

**One-line output:** `funding_line = "Saw you raised $52M in the Series B."` — the whole sentence
lives in the variable. §2's downstream gate explains why that matters more than it looks.

## 1. Trigger and scope

# Playbook: Google `site:` Keyword Filter

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** a campaign needs to know whether each company's own website talks about a specific
thing — a certification, a technology, a service line, a location, a program name — and no structured
API sells that field.

**Do not use when:** you need to *find* companies you do not have yet; when the thing you want is a
**page** rather than a phrase (`playbook-pricing-page`, `playbook-case-study-page`); or when a real
filterable field already exists (funding stage, headcount, technology installed) — filter at the
source and skip this entirely.

**One-line output:** `site_keyword_line = "you hold SOC 2 Type II certification"`, rendering as

# Playbook: Hiring Surge

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the angle is "you are scaling this team right now".

**Do not use when:** you want the job ad itself (`playbook-job-posting-language`), the person who
just started (`playbook-new-in-role`), or money rather than people (`playbook-fundraising`).

**One-line output:** `hiring_surge_line = "your sales team has 8 people who started new roles in the last six months"`

## The three things you cannot get wrong

1. **It is role starts, not hires. Measured 28.3% internal moves.** The filter counts people whose

# Playbook: Job Posting Language

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the angle is "you are hiring for this, and that is exactly what we help with". The
client names the roles or the description language that signals a fit, and every row needs a
specific, current opening to reference.

**Do not use when:** you want *how many* people they hired or headcount growth over time
(`playbook-hiring-surge`). If you want to **find** companies hiring a role rather than enrich a list
you already have, that is discovery mode — see note E.

**One-line output:** `job_posting_line_safe = "you're hiring an account executive for private equity"`

# Playbook: LinkedIn Engagement

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when** an account's audience is the client's audience — a **competitor** or the **client's own
customer**.

**Do not use for** more companies in a market, similar companies (`playbook-lookalikes`), or job
changers (`playbook-new-in-role`). The chain is the same for both lanes; only the target URL changes.

**Output:** one row per engager — who, employer, post, evidence URL. `engagement_line` is **opt-in**.

## The number that governs planning

# Playbook: Case-Study Lookalikes (filter mining, not raw lookalike)

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** a client has a case study, a flagship logo, or one obviously happy customer, and you
want the list of companies for whom that story is the strongest thing you could say.

**Do not use when:** you are building a whole market from an ICP sentence with no standout customer
to anchor on; or you want to name a customer off the **prospect's own** case-study page ("saw your
work with Intercom") — that is the opposite direction and belongs to `playbook-case-study-page`.

See also `disco-like` in this repo for plain seed-domain lookalike discovery. **This playbook is the
higher-precision version**: it mines *why* the story resonates before it searches.

# Playbook: Name 2 Other Prospects

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the email needs a "not sure if this is you" or "who owns this" angle, or any CTA that
offers the reader an easy referral instead of a yes or no.

**Do not use when:** you want the buying committee itself as separate leads. **You never email the
people named in this variable** unless they are separately in the list on their own merits.

**One-line output:** `other_prospects = "Jeff Barg or Rob Cook"`

## 1. Trigger and scope

# Playbook: New in Role

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the campaign angle depends on the buyer being new in the seat, because a new leader
is rebuilding their stack, has budget to reallocate, and has not yet formed a vendor preference.

**Do not use when:** you want people who moved to a **new company** regardless of title, or you
want the company-level hiring story. Use `playbook-hiring-surge` for headcount growth and
`playbook-fundraising` for the "they just raised, so they are buying" angle.

**One-line output:** `new_in_role_line = "you stepped into the COO seat at Northwind in April"`

## 1. Trigger and scope

# Playbook: Pricing Page

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the campaign needs to separate self-serve companies from sales-led ones, or needs a
real price point to anchor an offer against.

**Do not use when:** you want technology installed on the site (`playbook-tech-on-website`).

**One-line output:** `pricing_public = true` with
`plans = [{"name":"Basic","price":"$10","period":"month"}, ...]` and `lowest_paid_price = "$10"`.

## 1. Trigger and scope

# Playbook: Company Social Link Finding

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** a campaign or a downstream playbook needs the company's own social profile URL — most
often LinkedIn for engagement scraping, or Instagram and Facebook for a DTC brand angle.

**Do not use when:** you want a **person's** profile, post-level engagement
(`playbook-linkedin-engagement`), or an ad-library readout (`playbook-ad-library`, which consumes
this playbook's Facebook output).

**One-line output:** `company_linkedin_url = "https://www.linkedin.com/company/yampa-precision-manufacturing"`

## 1. Trigger and scope

# Playbook: Recent Social Post

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the campaign wants proof you read the account's content and there is no sharper
structural signal (funding, new-in-role, hiring surge).

**Not this playbook:** who *engaged* with a post (`playbook-linkedin-engagement`); a job change
(`playbook-new-in-role`).

**Output:** `social_post_text` + `social_post_url` + `social_post_days_ago` + `social_post_author_type`.
Optional, on request only: `social_post_line` → `Noticed they became an IBM Business Partner.`

## Six decisions that govern every run

# Playbook: Technology On Website

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

## 1. Trigger and scope

**Use when:** the offer only makes sense to companies running a specific piece of software on their
site — a Shopify app, a Klaviyo migration, a HubSpot implementation, a WordPress maintenance offer.
Works in both directions: **discovery** ("build me a list of every company on Shopify") and
**enrichment** ("here are 4,000 domains, which run Klaviyo").

**Do not use when:** you want the software a company *sells*, or mail-infrastructure targeting.

**Default output:** `tech_confirmed`, `tech_stack_verified` (JSON array), `tech_evidence`,

# Playbook: Warm Intros

> All rules here are best practice, not law. Override any of them when the campaign calls for it; note the best practice once and move on.

**Use when:** the campaign's whole premise is that the prospect is **already one hop from the
client**, and you need both the list and the line. The strongest version, and the one to build first,
is *"you used to work at a company that is now our client's customer"*.

**Do not use when:** you want a second contact at an account you already email
(`playbook-name-to-other-prospects`), people who engaged with a competitor's or customer's posts
(`playbook-linkedin-engagement`), or companies similar to a seed set (`playbook-lookalikes`).

**One-line output:** `warm_intro_line = "you came up through Coca-Cola, who we work with now"`,
rendering as `Noticed you came up through Coca-Cola, who we work with now.`

# Positive Reply Scoring

**Reply rate tells you if people are paying attention. Positive reply rate tells you if they want what you're selling.** This skill computes the second.

## Why this exists

A campaign can get 5% reply rate and still be a disaster. If 90% of those replies are "unsubscribe" and "not a fit," you're burning your domains for nothing.

The metric that matters is:
```
positive_reply_rate = positive_replies / total_sent
```

Compared side-by-side:

# Prospeo Full Search Export

Extract your entire Prospeo people search to a CSV file. Build your search in Prospeo's UI, then let Claude pull every single result via the API — even if the search has more than 25,000 results.

## Required step: Qualify with /icp-prompt-builder (do not skip)

Before exporting more than 500 contacts, run Prospeo on a 50-contact sample, then invoke `/icp-prompt-builder` to tune a qualification prompt in 3-5 rounds of 10 (with your approval each round). Apply the tuned prompt to the full export to filter out bad fits.

**Why required:** email enrichment downstream costs $0.05-$0.15 per person. A 25K export that's 40% wrong-fit wastes $500-$1,500 on email-finding that goes nowhere. The ICP prompt builder takes 10-15 min and saves that cost 40-70% of the time.

**Safe skip:** only if your Prospeo filter is already extremely tight (e.g., 5 exact titles + 1 industry + narrow headcount) AND you've run the same filter successfully before. Even then, run `/icp-prompt-builder` on 10 samples as a sanity check — it's nearly free to confirm.

## What This Does

# Prospeo Search Person API

This skill documents how to use the Prospeo Search Person API for finding leads with filters.

## When to Use

Use this skill when:
- Searching for people/leads by job title, industry, location, company size, etc.
- Building lead lists from Prospeo's database
- Running large US-wide searches that need state-by-state crawling

## API Overview

**Endpoint:** `POST https://api.prospeo.io/search-person`

# Smartlead API Reference

## Authentication

[Credential-related source omitted]

Environment variable: `$SMARTLEAD_API_KEY`

Base URL: `https://server.smartlead.ai/api/v1`

## Rate Limiting

- **Standard plan:** 2,000 req/min (33 req/sec)
- **Higher plans:** 3,000 req/min (50 req/sec)

# Smartlead Campaign Upload

Takes a `variants.yaml` (written by `/campaign-copywriting`) and a `leads.csv` (from your list-building skills) and creates a DRAFT campaign in Smartlead. You review the result in the Smartlead UI and press Start manually.

**This skill does NOT ship any email copy.** Copy comes from `/campaign-copywriting`. This skill is the mechanical upload layer — API calls only.

## Why DRAFT only

Cold email launches should never happen from a script. You hit Start in the Smartlead UI after reviewing:
- Subject lines + body previews
- Inbox assignments (correct tag? correct count?)
- Lead count and a few random lead rows
- Schedule (timezone + hours + throttle)

# Smartlead Inbox Manager

Zapmail hands you hundreds of new inboxes. Smartlead needs them configured: warmup enabled with the right ramp, signatures set so emails don't look bare, tags applied so you know which are active vs insurance, and health monitored so dead inboxes get recycled.

This skill does all of that via the Smartlead API.

## Core operations

| Operation | What it does | Script |
|---|---|---|
| Enable warmup | Turns on warmup with correct ramp (start 1/day, +5/day up to 40/day) | `set-warmup.ts --mode=enable` |
| Disable warmup | Turns off warmup (use for insurance inboxes) | `set-warmup.ts --mode=disable` |
| Set signatures | Bulk-applies a signature template to inboxes | `set-signatures.ts` |
| Tag as active | Adds the "active" tag | `tag-inboxes.ts --tag=active` |

# Smartlead Spintax Skill

Add deliverability-improving spintax to cold email sequences using Smartlead's syntax, without changing the meaning or tone of the original messaging.

## Core Concept

Spintax lets Smartlead randomly pick one option per send, so no two emails are identical. This reduces pattern detection by spam filters and improves deliverability.

**Smartlead syntax:** `{option1|option2|option3}`

Each option inside the curly braces is separated by a pipe `|`. Smartlead picks one at random per email send.

---

# Spam Word Checker

Always-on spam and deliverability rules. Apply to every subject line, opener, second line, follow-up, and CTA generated or reviewed.

---

## When These Rules Apply

- Subject lines, openers, second lines, follow-ups, CTAs
- Any cold outreach copy being written or QA'd
- If a deliverability checker or review flags a word/phrase, treat it as banned going forward unless user explicitly approves an exception
- When in doubt, choose the lower-hype rewrite

---

# Cold Email Domain Setup: Dynadot + Zapmail (Public Edition)

Complete workflow to generate short domain names, purchase them on Dynadot, and set them up on Zapmail with inboxes ready for cold email sending.

---

## First-Time Setup (Onboarding)

Before running anything, you need two API keys stored as environment variables.

### 1. Get Your Dynadot API Key
1. Log into [dynadot.com](https://www.dynadot.com)
2. Go to **Tools > API** in the left sidebar
3. Enable API access and copy your API key