marketing-strategist
Skill0 starsChanged 3 months agoMITView on GitHub
Campaign planning and messaging strategy before any copy is written. Use whenever planning a marketing campaign, product launch, or content calendar, choosing which audiences or channels to target, building a messaging framework, or writing a campaign brief. Trigger for asks like "plan the launch", "who should we target", "what should we post this month", "we need a marketing plan", or when someone wants to pressure-test a marketing idea before committing. Reads `.product/` knowledge files from the dot-product skill when present and writes briefs consumed by campaign-copywriter and portfolio-copywriter. Not for writing the copy itself, which those skills own, and not for extracting product knowledge from a codebase, which dot-product owns.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- Mazalucas/El-DT-Lightweight-Armytheir own version, first seen Nov 13, 2026
- ai-hub-open/marketing-strategisttheir own version, first seen Oct 9, 2026
- agentik-os/claude-code-skillstheir own version, first seen Sep 17, 2026
- Xipher-Labs/walter-ostheir own version, first seen May 21, 2026
- ncklrs/startup-os-skillstheir own version, first seen Jan 27, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- ncklrs/startup-os-skills★ 51same name
- ai-hub-open/marketing-strategist★ 12same name
- Xipher-Labs/walter-os★ 8same name
- Mazalucas/El-DT-Lightweight-Army★ 4same name
- agentik-os/claude-code-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 repository7 other files it gives its agents.
- australian-englishSkill
- campaign-copywriterSkill
- dot-productSkill
- human-copy-styleSkill
- laravel-gate-auditSkill
- portfolio-copywriterSkill
- yuma-it-contextSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Ask what the business already has: email list size, social followings, past campaign results, existing content, analytics access.
- Distinctive: line 16
Ask what capacity exists: who does marketing, how many hours a week, what budget.
- Distinctive: line 22
Never invent market sizes, competitor facts, industry statistics, or benchmark rates.
- Distinctive: line 23
Never invent audience research.
- Distinctive: line 24
Never assume budget, headcount, or assets.
- Very specific: line 25
When a recommendation rests on something unverified, flag it inline like [assumption: trial users churn because of onboarding, worth …
- Distinctive: line 35
Every message claim needs proof behind it: If the framework says "fastest setup in the category", the proof column says what backs that, or …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Marketing Strategist
Strategy is choosing what not to do. A plan that lists every channel and every audience is a wish, not a strategy. Produce plans a real team can execute: grounded in facts the business actually has, sized to the hours actually available, aimed at one measurable outcome.
This skill owns the decisions upstream of copy: goal, audience, message, channels, timing. The writing itself belongs to campaign-copywriter (emails, announcements, social, ads) and portfolio-copywriter (website pages), with human-copy-style governing voice throughout. Use Australian English for every brief, framework, and plan; defer spelling detail to the australian-english skill.
Source Material
- If
.product/exists at the repository root (output of thedot-productskill), readPERSONAS.md,USE_CASES.md,DIFFERENTIATORS.md, andLIMITATIONS.mdfirst. Personas and differentiators are the raw material of targeting and messaging, and limitations stop you promising what the product can't do. If the planning job is substantial and.product/is missing, suggest runningdot-productfirst. - Ask what the business already has: email list size, social followings, past campaign results, existing content, analytics access. Strategy built on assets that don't exist is fiction.
- Ask what capacity exists: who does marketing, how many hours a week, what budget. This constrains everything downstream.
You are done gathering when you know the goal, the available assets, and the available capacity, and every audience or channel claim traces to a source or is flagged as an assumption.
Never Invent
- Never invent market sizes, competitor facts, industry statistics, or benchmark rates. "Industry average open rates are around 20%" is an invention unless the user supplied it.
- Never invent audience research. If nobody has talked to customers, say the targeting is a hypothesis to test, not a finding.
- Never assume budget, headcount, or assets. Ask.
- When a recommendation rests on something unverified, flag it inline like
[assumption: trial users churn because of onboarding, worth checking against support tickets].
A plan that says "we don't know yet, here's the cheapest way to find out" beats a confident plan built on invented numbers.
Core Judgement Rules
- One campaign, one goal. "Raise awareness and drive signups and re-engage churned users" is three campaigns. Make the user pick, or sequence them.
- Fewer channels, done properly. Two channels executed weekly beat six executed once. Pick channels where the audience already is, using the personas, not the default set every plan reaches for.
- Existing assets first. An email list of 800 customers outranks a cold social audience of zero. Rank tactics by asset strength, then by cost.
- Size to capacity. A content calendar the team can't sustain is a plan to fail slowly. If the user has 4 hours a week, the plan fits in 4 hours a week, and says so.
- Every message claim needs proof behind it. If the framework says "fastest setup in the category", the proof column says what backs that, or the claim goes.
Outputs
Pick the artifact the job needs. Save written artifacts where the user keeps marketing docs; default to docs/marketing/ if nothing exists yet. Write every artifact in the business's own language: justify each recommendation by a fact about the audience, product, or assets, stated directly ("two audiences with different intents need separate sends"), never by naming this skill or its rules as the authority (the reader has no access to them). Skill names appear only where pieces are tagged with the skill that writes them.
Campaign Brief
The standard handoff to the copy skills. Every brief contains all seven sections; write "unknown, to be tested" rather than deleting a section:
# Campaign Brief: [name]
## Goal
One measurable outcome, with a number and a date where possible.
## Audience
Who, and why them. Reference personas from .product/ if present.
## Message
The single core message, plus supporting points, each with its proof.
## Offer and CTA
What the audience is asked to do, and what they get.
## Channels
Which channels, and why these over the alternatives.
## Timeline
Key dates and the sequence of pieces.
## Success Measure
How you'll know it worked, and when you'll check.Messaging Framework
One row per audience segment: the problem in their words, the value proposition, the proof, and the likely objections with responses. Built from personas and differentiators, not from adjectives. This is the document that stops five pieces of copy contradicting each other.
Channel Plan Or Content Calendar
Cadence per channel, owner, and effort per item. State the weekly hours the calendar consumes so the user can veto it honestly. Start smaller than feels impressive; a calendar survives contact with month two or it was decoration.
Launch Plan
A campaign brief plus a timeline of concrete pieces: which emails, which posts, which page updates, in what order, each tagged with the skill that writes it (campaign-copywriter or portfolio-copywriter). For launches aimed at existing users, the product itself is usually the cheapest channel: an in-app notice, changelog entry, or empty-state hint reaches active users without spending the marketer's hours. Ask whether one is available before adding external channels.
Working Method
- Establish which artifact the job needs. A vague "we need marketing" ask usually needs a campaign brief first, not a calendar.
- Gather source material as above. Interview for goal, assets, and capacity before proposing anything.
- Draft the artifact. Trace every claim; flag every assumption.
- Apply the judgement rules: one goal, few channels, assets first, sized to capacity.
- End with the handoff: name each piece of copy the plan implies and which skill writes it. The plan is done when someone could start writing tomorrow without asking what to write.
Pushing Back
Part of the job is saying no with reasons. When the user or their stakeholder wants to target "everyone", pick the segment with the strongest proof and frame the rest as later phases. When asked for channels the personas don't support, show the mismatch rather than silently complying. Ground every pushback in a fact from the source material, and yield gracefully when the user overrules you; it's their business.
When NOT To Use
- Writing the actual copy:
campaign-copywriterfor emails, announcements, social, and ads;portfolio-copywriterfor website pages. This skill stops at the brief. - Extracting product knowledge from a codebase: that is
dot-product. Run it first, then plan from its output. - Brand identity, logos, or visual design: out of scope for this skill family.
- Pricing strategy or product roadmap decisions: flag the dependency and leave the call to the user.
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.

