Kai launch
Open-source AI CMO for Claude Code: marketing agent skills for SEO, content, email, ads, launches, CRO, AEO/GEO, and AI-search visibility.
npx -y skills add cgallic/kai-cmo-harness --skill kai-launchAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
What its author says it does
Copied from the file, not written here
Plan and produce a complete product launch marketing package — landing page copy, email sequences, ad campaigns, press release, social posts, and launch timeline. Orchestrates all other kai skills into a coordinated launch. Use when "product launch", "launch campaign", "go-to-market", "GTM plan", "launch marketing", "we're launching", "prepare launch materials", or any request to coordinate marketing for a new product, feature, or major update.
SKILL.md
6.8 KB, ~1.6k tokens by cl100k_base, as published. Nobody here has run it
Orchestrate a complete product launch — coordinates emails, ads, content, PR, and social into a unified campaign with a timeline.
Phase 0: Load Product Context
Check if MARKETING.md exists in the project root (same directory as CLAUDE.md, README.md, package.json).
If it exists: Read it — skip product discovery questions. It has the product name, ICP, value prop, monetization, brand voice, current channels, and competitive landscape.
If it does NOT exist: Auto-explore the codebase to create it in the project root (next to CLAUDE.md). Do NOT ask the user what the product is. Read CLAUDE.md, README.md, PROJECT.md, package.json, landing pages, and any project files. Search for email/ad/analytics config. Then create MARKETING.md using the template from /kai-email-system. Present draft to user for confirmation.
Phase 1: Launch Discovery
Read from MARKETING.md. Only ask about things not covered there:
- What's launching? (new product, feature, major update, rebrand)
- Launch date — when does it go live?
- Channels available — email list size, ad budget, social following, press contacts?
- Offer/hook — any launch pricing, early access, beta invite?
- Landing page — exists or needs to be created?
Phase 2: Launch Timeline
Generate workspace/launch/_timeline.md.
Standard Launch Phases
| Phase | Timing | Activities |
|---|---|---|
| Pre-launch | T-14 to T-7 | Teaser emails, waitlist, social hints, internal prep |
| Warm-up | T-7 to T-1 | Blog post(s), detailed previews, influencer/press outreach |
| Launch Day | T-0 | Announcement email, ad campaigns go live, press release, social blitz |
| Post-launch | T+1 to T+14 | Nurture sequence, retargeting ads, case study collection, performance review |
| Sustain | T+14 to T+30 | Content marketing, SEO articles, ongoing ads optimization |
Asset Checklist
| Asset | Channel | Phase | Status |
|---|---|---|---|
| Landing page copy | Web | Pre-launch | Needed |
| Teaser email (1-2) | Loops | Pre-launch | Needed |
| Announcement email | Loops | Launch day | Needed |
| Follow-up email sequence (3-5) | Loops | Post-launch | Needed |
| Meta ads — TOF (3 variants) | Meta | Launch day | Needed |
| Meta ads — retarget (3 variants) | Meta | Post-launch | Needed |
| Google ads — brand (RSA) | Launch day | Needed | |
| Google ads — non-brand (RSA) | Launch day | Needed | |
| Blog post — announcement | Blog | Warm-up | Needed |
| Blog post — deep dive | Blog | Post-launch | Needed |
| LinkedIn article | Launch day | Needed | |
| Press release | PR | Launch day | Needed |
| Social posts (5-10) | Social | All phases | Needed |
Adapt to the actual launch. Remove assets for channels the user doesn't have.
Approval Gate
Present the timeline and asset checklist. Confirm before producing anything.
Phase 3: Batch Production
Produce assets in dependency order:
Order of Operations
- Landing page copy — defines the core messaging everyone else references
- Press release — crystallizes the announcement narrative
- Blog posts — expand on the story
- Email sequences — use
/kai-email-systemworkflow for the launch email set - Ad campaigns — use
/kai-ad-campaignworkflow for the ad set - Social posts — extract key lines from above assets
- LinkedIn article — repurpose blog post for LinkedIn format
Per-Asset Production
Each asset follows the standard harness pipeline:
- Load the right framework from
knowledge/ - Load the skill contract from
harness/skill-contracts/ - Load platform policy (for ads) from
harness/references/ - Write against the framework + persona
- Run quality gates
- Max 2 retries per asset
Messaging Consistency
All assets must use the same:
- Core value proposition (defined in landing page copy)
- Key stats/proof points (consistent across all channels)
- CTA (same destination — landing page or signup)
- Persona hooks (same pain points, same language)
Extract these from the landing page copy and use as a reference for all subsequent assets.
Batch Output
workspace/launch/
├── _timeline.md
├── _messaging-guide.md # Core VP, stats, CTA, extracted from landing page
├── landing-page/
│ └── copy.md
├── emails/
│ ├── teaser-1.md
│ ├── teaser-2.md
│ ├── announcement.md
│ ├── follow-up-1.md
│ └── follow-up-2.md
├── ads/
│ ├── meta/
│ └── google/
├── blog/
│ ├── announcement.md
│ └── deep-dive.md
├── pr/
│ └── press-release.md
├── social/
│ ├── launch-day-posts.md
│ └── sustain-posts.md
├── linkedin/
│ └── article.md
└── _quality-report.md
Phase 4: Quality Report
# Launch Campaign Quality Report
> **Kai root note:** `knowledge/`, `harness/`, and `scripts/` paths in this skill live in the Kai install, not the user's project. Resolve them against the first ancestor directory of this SKILL.md that contains a `knowledge/` folder (the Kai plugin root, `~/.claude/kai`, or the kai-cmo-harness repo). `MARKETING.md`, `memory/`, and any output files live in the current project. If a referenced `scripts/` command is not available in this install, say so, skip it, and continue with the file-based guidance — never fabricate its output.
## Summary
- Total assets: [N]
- Passed all gates: [N]
- Channels covered: [list]
- Messaging consistency: [PASS/FAIL — same VP across all assets?]
## Per-Asset Results
| Asset | Channel | Four U's | Banned | Policy | Status |
|-------|---------|----------|--------|--------|--------|
## Launch Readiness
- [ ] Landing page copy approved
- [ ] Email sequences loaded in Loops
- [ ] Ad campaigns ready to activate
- [ ] Press release ready to distribute
- [ ] Blog posts scheduled
- [ ] Social posts queued
Phase 5: Post-Launch Monitoring Plan
Generate workspace/launch/_monitoring.md:
- Day 1, 3, 7, 14 check-in schedule
- Metrics to watch per channel (email open rates, ad CTR, landing page conversion)
- When to adjust (kill underperforming ads, boost winners)
- Content to produce based on early results (FAQ post, case study, feature tutorial)
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most product growth skills give in ~1.6k tokens
Counted across 728 of the 1,010 authors here whose files we hold, read 2026-08-07
- Read product marketing context before asking questionsin 24 of 728, across 18 files
- Define the ideal customer profilein 21 of 728, across 3 files
- Document a rollback plan before deploymentin 21 of 728, across 12 files
- Analyze the codebase to understand the productin 19 of 728, across 1 file
- Ask clarifying questions about the value propositionin 19 of 728, across 1 file
- Search for companies matching the criteriain 19 of 728, across 1 file
- Look for signals of immediate needin 19 of 728, across 1 file
- Assign a fit score from one to tenin 19 of 728, across 1 file
- Identify the target decision-maker rolein 19 of 728, across 1 file
- Suggest a personalized contact strategyin 19 of 728, across 1 file
- Provide conversation starters for outreachin 19 of 728, across 1 file
- Format results in a scannable markdown templatein 19 of 728, across 1 file
Said here and by no other author read
- read MARKETING.md if it exists
- create MARKETING.md if it is missing
- do not ask what the product is
- ask only what MARKETING.md does not cover
- confirm the timeline before producing assets
- produce assets in dependency order
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.