marketing-context

Overview of marketing-context skill

What marketing-context creates

marketing-context is a Product Marketing setup skill for creating the shared context file that other marketing skills can read before they write copy, messaging, SEO pages, launches, or sales assets. Its main output is .claude/product-marketing-context.md, a canonical document covering product overview, ICP, positioning, differentiation, customer language, brand voice, proof points, objections, and related marketing constraints.

Best fit for Product Marketing teams

The marketing-context skill is most useful when you are starting a new product marketing workspace, onboarding Claude to an existing product, or tired of restating your audience, tone, competitors, and positioning in every prompt. It fits founders, PMMs, growth marketers, content strategists, and agencies that need repeatable context before using downstream marketing skills.

Why this skill is different from a normal prompt

A generic prompt can ask for “brand voice” or “ICP,” but marketing-context turns that into a reusable operating document. The skill supports three practical modes: auto-drafting from repository or website materials, guided interview when source material is thin, and updating an existing context file as positioning evolves. It also includes a validator script to score completeness, which helps prevent a vague, half-filled marketing brief from becoming the default source of truth.

Adoption considerations

Install marketing-context when you want consistency across multiple AI-assisted marketing tasks. Do not expect it to invent reliable positioning from nothing; it needs source material, interviews, or user corrections. The most important setup detail is path discipline: sibling skills in this library expect the context at .claude/product-marketing-context.md, not a random root-level note.

How to Use marketing-context skill

marketing-context install and files to inspect

Install from the GitHub skill repository with:

npx skills add alirezarezvani/claude-skills --skill marketing-context

After install, read SKILL.md first to understand the workflow and required output path. Then inspect templates/marketing-context-template.md to see the expected sections and scripts/context_validator.py to understand how completeness is scored. If you already have .agents/marketing-context.md or marketing-context.md, move or consolidate it into .claude/product-marketing-context.md so other marketing skills can find it.

Inputs that produce a useful first draft

For auto-draft mode, point the assistant to real materials: README.md, landing pages, pricing pages, docs, customer quotes, case studies, sales decks, competitor notes, and support tickets. A strong request is specific about what the assistant may infer and what must be flagged as unknown.

Example:

“Use marketing-context to create .claude/product-marketing-context.md. Review README.md, docs/, website/, and pricing.md. Draft a V1 for a B2B SaaS product selling to RevOps teams. Mark uncertain claims as assumptions, list gaps at the end, and do not invent customer proof points.”

This gives the skill enough boundaries to build context without turning speculation into messaging.

Turning a rough goal into a complete prompt

Weak prompt: “Set up marketing context for my app.”

Better prompt: “Use marketing-context for Product Marketing setup. I need a reusable context file for future landing page, email, and SEO skills. Interview me one section at a time if details are missing. Prioritize ICP, jobs to be done, competitive alternatives, differentiation, customer language, brand voice, and proof points. Save the result to .claude/product-marketing-context.md.”

This works because it names the downstream use cases, required sections, interview style, and save location.

Suggested workflow for first-time usage

Start with auto-draft if your repo or website already describes the product. Review the draft as a PMM, not as a copy editor: fix audience, category, competitive framing, claims, and “words to avoid” before polishing tone. Then run or reference scripts/context_validator.py to identify missing sections. Finally, ask the assistant to update the context after meaningful changes such as a new ICP, new pricing, new competitor, new proof point, or repositioning.

marketing-context skill FAQ

Is marketing-context only for technical products?

No. The repository path sits inside a marketing skill library, and the template works for SaaS, marketplaces, services, e-commerce, and other product types. Technical repositories make auto-draft mode easier because the assistant can inspect docs and code-adjacent files, but non-technical teams can use guided interview mode with business inputs instead.

What does marketing-context need before install?

You do not need a perfect positioning strategy before installing. You do need willingness to provide corrections. Minimum useful inputs include what the product does, who buys it, the main use case, alternatives customers compare against, why customers choose you, preferred tone, and any proof you can safely claim.

When should I not use marketing-context?

Do not use marketing-context as a substitute for customer research, legal review, brand strategy, or competitive intelligence. It organizes and operationalizes what you know; it does not verify market truth. It is also overkill for a one-off social post if you will not reuse the context across additional marketing work.

How does it fit with other marketing skills?

marketing-context should run before downstream Product Marketing, content, SEO, launch, or copywriting skills. Its value compounds when multiple skills read the same .claude/product-marketing-context.md, because each output starts from the same ICP, positioning, voice, proof points, and constraints.

How to Improve marketing-context skill

Strengthen the source material before rerunning

The best marketing-context results come from concrete evidence, not broad claims. Add real customer phrases, sales objections, support pain points, competitor names, pricing notes, screenshots of current positioning, and approved proof points. If you only provide “AI productivity tool for teams,” expect generic output. If you provide “mid-market RevOps teams replacing spreadsheet-based territory planning because handoffs break after comp changes,” the skill can produce sharper positioning.

Use the validator as an editorial checklist

context_validator.py scores sections such as Product Overview, Target Audience, Problems & Pain Points, Competitive Landscape, Differentiation, Customer Language, Brand Voice, and Proof Points. Treat a low score as a signal to improve the brief before asking other skills to write. The point is not to chase a perfect number; it is to catch missing ingredients that will weaken every downstream asset.

Common failure modes to correct

Watch for invented proof points, vague personas, competitor lists without “how they fall short,” brand voice adjectives without examples, and differentiation that describes features rather than buyer-relevant advantage. Ask the assistant to label assumptions, separate facts from hypotheses, and add a “needs validation” section when evidence is thin.

Iterate marketing-context after the first output

Your first context file should be a working V1, not a final brand bible. After reviewing it, run targeted update prompts: “Update only Competitive Landscape and Differentiation based on these new competitors,” or “Revise Customer Language using these five sales call excerpts.” This keeps marketing-context accurate without rewriting stable sections every time.