marketing
Skill1 starsChanged 23 days agoMITView on GitHub
A marketing plan one founder can actually run. Sets the positioning from the rivals and the objections (not adjectives), picks the channels these buyers already use, computes the most you can pay to get a customer with a small tool (math shown), and writes a dated 30-day plan paced for one person, with what to post, send or do each day and the number that says keep, change or stop. Local businesses get local channels (offices, walk-ins, partners, map listings, neighborhood groups). The founder posts and sends everything. Use it when someone says "marketing plan", "how do I get customers", "where do I find my first customers", "what should I post", "30 day plan", "how much can I spend on ads", "customer acquisition cost", "positioning", or "nobody knows we exist". Not for ad account management or writing one email (biz-boost skills do that).
32 other projects use a version of this.
Adopted by
Most recently taken up by
- stericishere/skillstheir own version, first seen Nov 15, 2026
- muthuishere/toolnexustheir own version, first seen Nov 8, 2026
- fal-ai-community/skillstheir own version, first seen Nov 7, 2026
- FradSer/skillstheir own version, first seen Nov 1, 2026
- mxrsv/agents-skillstheir own version, first seen Nov 1, 2026
- HKTITAN/isb-skillstheir own version, first seen Nov 1, 2026
Used with
Projects with it also use writing, research and 4 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
alxyrginSkillwriting
Создание контента для личного бренда — статьи, посты, туториалы. Use when user asks to "напиши статью", "создай пост", "подготовь контент", "write article", "create content", or when task requires content creation for blogging, LinkedIn, Medium, or other publishing platforms.
Also in 3 projects that use this
NVIDIASkillresearch
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
Also in 4 projects that use this
GoogleSkillprototype
Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.
Also in 3 projects that use this
BlockSkillfrontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
Also in 4 projects that use this
SentrySkillskill-creator
Create new agent skills following the Agent Skills specification. Use when asked to "create a skill", "add a new skill", "write a skill", "make a skill", "build a skill", or scaffold a new skill with SKILL.md. Guides through requirements, writing, registration, and verification.
Also in 3 projects that use this
GoogleSkillcode-review
Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.
Also in 3 projects that use this
Other versions32 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 32
- fal-ai-community/skills★ 265same name
- zeuikli/claude-code-workspace★ 155same name
- travisjneuman/.claude★ 106same name
- pantalytics/odoo-mcp-pro★ 92same name
- cgoinglove/thursday★ 56same name
- cgoinglove/thursday-agent★ 56same name
- arturseo-geo/marketing-skill★ 35same name
- echris6/ai-business-skills★ 16same name
- mxrsv/agents-skills★ 11same name
- sukirman1901/frontend-expert★ 9same name
22 more not listed
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 repository11 other files it gives its agents.
- boardSkill
- brandSkill
- competitorsSkill
- customer-discoverySkill
- fundingSkill
- launchSkill
- launch-videoSkill
- offerSkill
- opsSkill
- pitchSkill
- validateSkill
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 23 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
- Distinctive: line 38
Write one short statement in this order: for <who>, who <the situation>, <business> is the <category> that <the one difference>, unlike …
- Very specific: line 88
Start on the date given, or founder/launch.md minus 14 days, or tomorrow.
- Distinctive: line 99
Every give-away in the plan (a free meal per referral, a free week, samples) gets its cost next to it, from the unit cost.
- Very specific: line 102
Write two or three example messages for the top channel: a post, a short outreach note, a flyer line.
- Distinctive: line 141
No invented stats, testimonials, reviews, results or retention numbers.
“Common” is measured: kept by at least 6 in 10 of the 23 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
marketing
Most early marketing fails for one of three reasons: the message is about the product instead of the buyer, the founder is on channels their buyers do not use, or they spend more to win a customer than the customer is worth. This skill fixes all three, then gives the founder a month they can actually keep up.
What the user typed after /marketing: $ARGUMENTS
Reads and writes
- Reads first: the business context (
founder/context.md, else the active file under~/.founder/businesses/), for tone and words to avoid. Thenfounder/idea.md,founder/competitors.md,founder/offer.md,founder/pricing.md,founder/numbers.json,founder/cfo.md,founder/discovery/evidence.md,founder/panel/results.md,founder/launch.md(the launch date). List what exists. - Writes:
founder/marketing.md. - Tool:
scripts/max_cac.pycomputes the most you can pay for one customer.
Step 1: positioning
Write one short statement in this order: for <who>, who <the situation>, <business> is the
<category> that <the one difference>, unlike <what they use now>, which <its weakness, from
the complaints>.
Take "what they use now" and "its weakness" from founder/competitors.md and the
objections. If those files are missing, ask the founder who they lose buyers to, and mark
the statement assumed until /competitors runs. A position nobody else can say is worth
more than a clever line.
Then three message lines, each answering one top objection from founder/offer.md or the
panel, in the buyer's words.
Step 2: the most you can pay for a customer
Run:
python3 -I "${CLAUDE_SKILL_DIR}/scripts/max_cac.py" --numbers founder/numbers.json --per-month <units a month> --months <months a customer stays>or with --price and --unit-cost if there is no numbers file. Use --churn for a
subscription if the founder knows it. It shows each step: contribution per unit, value per
customer, the break-even cost to win one, a safer target (a third of value by default), and
a payback cap if asked.
How long a customer stays and how often they buy are usually guesses at this stage. Say so,
use the founder's guess, and show what happens if it is half as long. If there is no
number at all, write the formula with {{...}} and point to /founder-cfo. Never invent a
retention rate.
Why it matters: the founder needs this number before any paid test, so they know when to stop a channel that is too expensive.
Step 3: pick the channels
List where these buyers already spend attention. Think about the buyer's day, not the platforms you know. Then pick two or three to test, not ten. For each: why these buyers are there, what a cheap test looks like, how many hours a week it takes, and what it costs.
| business | channels to consider first |
| local (food, services, shops) | the office or building manager, walk-in visits, partner businesses, neighborhood and workplace groups, the map listing, flyers where buyers wait, local events |
| online consumer | communities where the problem is discussed, short video showing the moment it works, search for the problem, creators the buyers follow, referrals |
| B2B | direct outreach to a named list, the communities the role belongs to, partners who already sell to them, content that answers the buyer's questions |
One person has about {{hours a week}} for marketing. Ask, and fit the plan to that number. A plan that needs 30 hours a week from someone with 8 gets dropped by day 5.
Step 4: the 30-day plan, with dates
Start on the date given, or founder/launch.md minus 14 days, or tomorrow. Write it in
four weekly blocks with real dates. For each day or task: what to do, which channel, who
does it (usually the founder), and the number to check. Put the first test of each channel
in week 1, so week 3 can double down on what worked.
At the end of each week, one line: the number to check and the level that says keep, change or stop. Derive the level from the max cost per customer in step 2 and write the math: "stop if a sign-up costs over $39.60, the safer target at half the stay". For a free channel, count the founder's hours at {{your hourly value}} or the cost of what is given away (samples, free meals, credits). A level with no reason behind it is just a guess.
Every give-away in the plan (a free meal per referral, a free week, samples) gets its cost next to it, from the unit cost.
Write two or three example messages for the top channel: a post, a short outreach note, a
flyer line. The founder sends and posts them. Run them through the humanize skill if it
is installed. If a draft is a cold email, include the sender's physical address and an
opt-out line (CAN-SPAM). No made-up stats, results or testimonials in the copy.
Step 5: save
Write founder/marketing.md:
# Marketing: <business>
Written: YYYY-MM-DD. Read: <files>. Assumed: <what was a guess>.
## Positioning
## Messages (one per top objection)
## The most you can pay for a customer
<tool output, and the "half as long" line>
## Channels to test
| channel | why these buyers are there | test | hours a week | cost |
## 30 days (YYYY-MM-DD to YYYY-MM-DD)
### Week 1 (dates)
...
## Keep, change or stop
| channel | number | keep if | stop if |
## Example messagesTell the user
The positioning in a line, the max cost per customer, the two or three channels, and what
to do on day 1. Next: /brand if the name and look are not set, /launch for the dated
launch, or the biz-boost skills (reach-plan, lead-gen, local-outreach) and the
LinkedIn or Instagram plan skills to go deeper on one channel.
Rules
- The founder posts, sends and calls. Never automate posting, messaging or scraping on any platform, and never buy or build contact lists against a site's terms.
- No invented stats, testimonials, reviews, results or retention numbers.
- Paid tests show their budget and wait for the founder's yes. This skill spends nothing.
- Cold email drafts carry a physical address and opt-out.
- Claims about health, money or results need proof; flag them for a lawyer if regulated.
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.

