agentleFS
JakeschincariolJakeschincariol/whale-skill/whale-hunter/SKILL.md

whale-hunter

Skill15 starsChanged 7 days agoMITView on GitHub

Whale Hunter. Finds the decision makers and pitches them. Looks up who signs the check, who owns the budget and who feels the pain at a company you want as a customer, using public business sources only, then writes a short pitch to the right person: what they pay today, the demo, and one small ask. Every email waits for your yes and goes out from your own account, with the legal footer and a plain way to opt out. Use when the user says "who decides at X", "find the decision maker", "find the CFO", "pitch them", "write the pitch", "cold email the buyer", "go after this company", or "hunt whales".

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Add people and deals with the shared board:

    line 28
  2. Very specific:

    Find the people: Search the company's own site (team, leadership, contact pages), press releases and professional profiles they published …

    line 45
  3. Very specific:

    If not, it uses what the market pays, and says "often".

    line 52
  4. Very specific:

    Log: Only after it went out: hunter.py record --approval "<their yes, word for word>".

    line 58
  5. Distinctive:

    Never sends, and never records an email that did not go out.

    line 70
  6. Distinctive:

    Never guesses an email address, and never emails anyone on the suppression list or marked do-not-contact.

    line 71
  7. Distinctive:

    Never asks for money in a cold email.

    line 72
  8. Distinctive:

    Never invents a fact about the company, a mutual friend, a past conversation or a fake "Re:".

    line 73
  9. Distinctive:

    Every cold email is honest and legal: a real subject line, the sender's real name and business, a postal address, and a plain way to opt …

    line 79
  10. Distinctive:

    Never invent a number, a result, a client or a deadline.

    line 84

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

whale-hunter

One big customer changes everything, and big customers don't come from ads. They come from a short, specific note to the one person who can say yes. The Hunter finds that person and writes that note. It never sends anything itself.

${CLAUDE_SKILL_DIR} below is this skill's folder (where this SKILL.md lives).

python3 "${CLAUDE_SKILL_DIR}/hunter.py" who --company north-pier
python3 "${CLAUDE_SKILL_DIR}/hunter.py" draft --deal north-pier --demo https://yourproduct.example/demo
python3 "${CLAUDE_SKILL_DIR}/hunter.py" record --deal north-pier --subject "You pay \$6,400 a month to chase invoices" \
    --text-file /tmp/pitch.txt --approval "yes" --at 2026-10-26T09:04

Add people and deals with the shared board:

python3 "${CLAUDE_SKILL_DIR}/../market-researcher/board.py" company --id north-pier --name "North Pier Logistics" --website https://northpierlogistics.com
python3 "${CLAUDE_SKILL_DIR}/../market-researcher/board.py" person --id renee-alvarez --company north-pier --name "Renee Alvarez" --title CFO \
    --decides signs --email renee@northpierlogistics.com --email_source https://northpierlogistics.com/team
python3 "${CLAUDE_SKILL_DIR}/../market-researcher/board.py" deal --id north-pier --company north-pier --person renee-alvarez --title "Founding customer"

Who decides

  • signs signs the check. A CFO or an owner. Pitch this person first.
  • budget owns the budget. A VP. Pitch them if the signer is out of reach.
  • pain feels the pain every day. A manager. They rarely buy, but they tell you what it's worth, so talk to them first.

Steps

  1. Find the people. Search the company's own site (team, leadership, contact pages), press releases and professional profiles they published for work. Save each one with board.py person and where you found them.
  2. Find an address they publish. --email_source is required. If the company doesn't publish one, use the general address or call. Never guess a pattern like first.last@company.
  3. Pick one. hunter.py who. One person at a company at a time: emailing three executives at once looks like spam, and they talk to each other.
  4. Draft. hunter.py draft. If you have a real number for this company (they published it or told you), it opens with it. If not, it uses what the market pays, and says "often". Fill any [fill: ...]. Make the first line more specific if you can, with something true. Under 90 words.
  5. Ask. Show the email. "Send this to Renee Alvarez at North Pier Logistics? Reply yes or no." One email per question, every time.
  6. Send. On a yes, create it as a Gmail draft from the user's account, and the user presses Send (or, if the user asked for it, send it through the Gmail connector right after that yes).
  7. Log. Only after it went out: hunter.py record --approval "<their yes, word for word>". /follow-up takes over. At most 50 pitches a day: it protects the user's inbox reputation and keeps this one to one, not bulk mail.

Example from the reel

North Pier Logistics: Renee Alvarez, CFO (signs the check); Tom Whitlock, VP Finance (owns the budget); Priya Nair, AR Manager (feels the pain daily). The pitch to Renee, subject "You pay $6,400 a month to chase invoices": "Hi Renee, North Pier pays about $6,400 a month to chase late invoices. We built a demo that does it automatically. Can I show you in 2 minutes?" (Illustrative: North Pier and everyone at it are fictional. On the example board the figure came from Priya, on a call.)

What it never does

  • Never sends, and never records an email that did not go out.
  • Never guesses an email address, and never emails anyone on the suppression list or marked do-not-contact.
  • Never asks for money in a cold email. The first ask is 2 minutes.
  • Never invents a fact about the company, a mutual friend, a past conversation or a fake "Re:".

Consent and truth

  • Nothing is sent, booked or charged without the user's explicit yes for that exact message. Claude drafts, the user says "yes", then it goes out from the user's own accounts. One yes per message. Never "yes to all".
  • Every cold email is honest and legal: a real subject line, the sender's real name and business, a postal address, and a plain way to opt out (CAN-SPAM in the US; in the UK, EU and Canada cold email to people needs a lawful basis or consent first, so check before emailing there). Opt-outs are honoured forever.
  • Public business information only: a company's own site and what people publish for work, never a person's private life, home or family, and nothing behind a login. A figure about a company's spending only goes in a pitch if they published it or told you.
  • Never invent a number, a result, a client or a deadline.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.