signal
Skill3 starsChanged 12 days agoMITView on GitHub
Your marketing director. Turns the buyer panel and the competitor map into positioning, the channels the target customer actually uses, a 30-day launch campaign with dated posts and ads, the first ten hooks, a budget split and the numbers to watch. Every message answers an objection real or simulated buyers raised. Use when the user says "how do I market this", "marketing plan", "launch campaign", "how do I get customers", "where do I advertise", "write my ads", or "positioning".
5 other projects use a version of this.
Adopted by
Most recently taken up by
- mia-cx/skillstheir own version, first seen Nov 8, 2026
- hiroqt/overdrivetheir own version, first seen Nov 1, 2026
- ai-ecoverse/skillstheir own version, first seen Oct 2, 2026
- envisioning/skillstheir own version, first seen Sep 22, 2026
- abouchard11/midnight-seo-skillstheir own version, first seen Aug 31, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- mia-cx/skills★ 4same name
- envisioning/skills★ 2same name
- hiroqt/overdrive★ 2same name
- abouchard11/midnight-seo-skills★ 1same name
- ai-ecoverse/skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- burn-testSkill
- dealbreakerSkill
- first-dollarSkill
- ground-truthSkill
- handshakeSkill
- kill-or-buildSkill
- litmusSkill
- prove-itSkill
- reality-checkSkill
- receiptsSkill
- shark-boardSkill
- straight-talkSkill
- truth-serumSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 20
Read ${CLAUDE_SKILL_DIR}/../receipts/rules.md first.
- Very specific: line 40
If none of these exist, say what is missing and offer to run /truth-serum --quick first.
- Distinctive: line 95
Write the kill line too: the number after 30 days that means the channel, or the idea, is not working.
- Distinctive: line 100
No fake reviews, follower counts, testimonials, before/afters or "as seen on".
- Distinctive: line 105
Treat buyers the way you would want to be treated: no pressure, no tricks, no talking about them as targets.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
signal
A business no one hears about does not exist. This skill plans how the first customers hear about it, using what the rest of the pack already learned instead of a generic playbook.
Read ${CLAUDE_SKILL_DIR}/../receipts/rules.md first. Every claim in
public copy must be one the ledger allows for a public audience:
python3 "${CLAUDE_SKILL_DIR}/../receipts/ledger.py" gate --audience publicStep 1: read the evidence
founder/evidence/calls/andfounder/evidence/claims.md: what real people said. Real words beat panel words; use them first.founder/panel/results.md: who buys, who doesn't, and why (simulated). The segments with the highest buy rate are the first audience to test, not a known audience. The top objections are what the marketing must answer.founder/competitors.md: how competitors position themselves (so you do not sound like them) and the gap.founder/offer.mdandfounder/pricing.md: what is being sold, at what price.founder/numbers.json: what a customer is worth, which caps what you can spend to win one.
If none of these exist, say what is missing and offer to run /truth-serum
--quick first. Do not invent an audience.
Step 2: positioning
One line, then the reasons:
For <the segment that buys most>, who <their problem in their words>,
<business> is the <category> that <the gap>, unlike <the nearest alternative>.Give three versions, recommend one, and say which panel evidence supports it.
Step 3: channels
Where comes before what. A plain ad in the place the buyers already are beats
a brilliant one in the wrong place (Ready, Fire, Aim, rule 11 in rules.md). Start
from where competitors keep advertising (founder/competitors.md, "how they
sell"): copy the medium that works, then be different in one thing. Do not
spend on making the company known (a brand campaign, a brochure, a launch
party); sell the product.
List where the first audience actually spends attention, specific to the business type and place: foot traffic and window signage, Google Maps, Instagram and TikTok, local creators, neighbourhood groups, partnerships with nearby businesses, email, paid search, paid social. For each: why it fits this buyer, rough cost, and how you will know it worked. Pick two or three to start; say what you are not doing and why.
Step 4: the 30-day campaign
founder/marketing.md gets a dated calendar from launch minus 14 days to launch
plus 16: what goes out each day, on which channel, with the message. Include:
- Ten hooks, each answering one panel objection or using one buyer reason, written for the format (a 3-second opening line for video, a headline for an ad, a sign in the window).
- The main ad on four legs: one big idea (a single idea, not five),
the big benefit to the buyer (about them, not the company), the big
promise that joins the two in one line, and the proof. Proof comes
only from claims
ledger.py gate --audience publicprints; if there is none yet, the ad says what the buyer can try at no risk instead of claiming results. - Two versions to test against each other in week one (two headlines, two offers or two prices), with the number that picks the winner.
- The launch-week offer from
founder/pricing.mdand when it ends. - What to post when there is nothing new to say (behind the scenes, the making of).
Step 5: budget and numbers
- The budget split across the chosen channels.
- The most you can pay to win a customer: "What one customer is worth" in
founder/cfo.md(first order plus the back end). Give marketing one target: a number of new customers by a date, each won for less than that. - The three numbers to watch weekly and the level that means "change something". Write the kill line too: the number after 30 days that means the channel, or the idea, is not working.
Rules
- No fake reviews, follower counts, testimonials, before/afters or "as seen on".
- Panel quotes are research; never use them as customer quotes in ads.
- A result the business has not delivered yet ("3x faster", "save 15 hours")
goes in copy as a promise or a goal with its terms, never as a result.
Nothing
PLANNEDorSIMULATEDis written as proof. - Treat buyers the way you would want to be treated: no pressure, no tricks, no talking about them as targets.
- Disclose paid partnerships. Follow the advertising rules for claims in the user's market (health and earnings claims especially). Not legal advice.
Output
founder/marketing.md, opening with The hard truth: the biggest reason the
first audience may not hear or care. Next: /straight-talk, then
/prove-it.
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.

