How to use the Product Launch Planner agent skill

What is an agent skill, and what does the launch skill make the agent do?

An agent skill is a folder with instructions (a SKILL.md file, plus optional reference files) that an AI agent loads when a request matches the skill's description. The agent then follows that guidance instead of answering from general knowledge alone.

The launch skill applies when you are preparing to release something publicly. It first checks whether the thing is ready to launch, using a Simple, Lovable, Complete (SLC) test: one clear job, something people would choose to use, and a whole experience with no glaring stubs. If it passes, the agent plans launch activities across three channel types: owned channels such as email, your blog, a community, and your website; rented channels such as social media and app stores; and borrowed channels such as partnerships, guest content, and speaking. It also works through a five-phase approach from internal launch to public release.

• Readiness check before any launch mechanics
• ORB planning across owned, rented, and borrowed channels
• Five-phase launch approach (internal, alpha, beta, and later stages)
• Related tasks: pre-launch QA, changelogs, release notes, and go-live checklists

How do you install the launch skill?

Download the ZIP for the skill from the directory page and unzip it. Put the resulting folder in one of these locations:

For a personal setup available across your projects, place it in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder so it applies only there.

If you use the Claude apps instead, upload the skill under Settings → Capabilities → Skills.

• Personal: ~/.claude/skills/
• Single project: .claude/skills/ inside the project folder
• Claude apps: Settings → Capabilities → Skills, then upload

How do you trigger the launch skill with natural requests?

You don't need a special command. The skill is selected from what you ask, so describe the release in plain words. The skill's description lists trigger phrases such as 'launch', 'Product Hunt', 'feature release', 'beta launch', 'early access', 'waitlist', 'go-to-market', 'pre-launch QA', 'launch checklist', 'changelog', 'release notes', and 'what did we ship this week'.

Good requests include the thing you're releasing and your situation. The more context you give, the more specific the plan. The skill also checks for a product marketing context file (for example .agents/product-marketing.md or .claude/product-marketing.md) and reads it first if present.

• 'We're about to ship a new reporting feature. How do I launch this?'
• 'Build a go-to-market plan for our beta launch next month.'
• 'Draft a launch checklist and release notes for version of our invoicing module.'
• 'What did we ship this week? Write a changelog for customers.'

What does the skill do for changelogs and release notes?

For shipped code, the skill's reference on turning shipped changes into marketing suggests an agent with repo access can read what was merged, decide what's worth telling customers, and draft it in customer language. Useful inputs include merged PRs with their descriptions, commits merged without a PR, changed files where a title is unclear, existing release notes or tags, and feature flags.

The guidance is explicit that only things customers can use today should be marketed. Anything behind a feature flag that is not live for customers should be left out.

• Typical window: the last week, or since the last release tag
• Commit log command the reference uses: git log --since="1 week ago" --no-merges --pretty=format:"%h %s (%an, %ar)"

How do you run a pre-launch and site-launch QA check?

The skill includes a site launch QA checklist for the moments before and after a website goes live: a new site, a redesign, a platform migration, or a major new section. It runs in three passes: before the switch, at the switch, and in the first two weeks after.

Checks before the switch include confirming every planned page has final copy (no lorem ipsum, TBD, or test prices), crawling staging for broken internal links and missing images, confirming legal pages are present, and submitting every form with a test entry to confirm it arrives where it should.

• Before the switch: content, links, forms, and conversion tracking
• At the switch: go-live checks
• First two weeks: post-launch monitoring

Examples of what the skill produces

Ask for a launch plan and the agent works through the readiness gate first. If the product fails it, expect advice to cut scope until one thing is lovable and complete rather than adding more. If it passes, expect a phased plan that names owned, rented, and borrowed channel actions.

For a Product Hunt or feature announcement, the skill points toward turning visibility into owned relationships, for example sending social traffic to a landing page with a signup form or email onboarding. The source's examples include an invite-only waitlist with one-on-one onboarding and a reviewer who was sent a free product rather than a paid sponsorship.