Skip to content

cgallic/kai-cmo-harness / harness/skills-v2/kai-growth-plan

kai-growth-plan

Verified

Generate a stage-appropriate marketing plan based on your company's MRR/stage. Uses the marketing-by-stage playbook to tell you exactly what to do (and what NOT to do) at pre-launch, early ($0-10K MRR), growth ($10-100K MRR), or scale ($100K+ MRR). Use when "what should I do for marketing", "growth plan", "marketing plan", "I just raised a round", "marketing strategy", "what's the right marketing for my stage", "GTM strategy", or any request for a stage-appropriate marketing roadmap.

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.

Objective

For reusable prompt selection, non-SaaS businesses or hybrid offers, load harness/references/marketing-growth-library/README.md, its catalog and R0 in prompts.md. Establish the business profile, outcome unit, evidence and permissions before applying the stage model below. MRR bands apply to recurring software revenue, not every business. Follow only relevant P1–P23/C1–C6 prompts and their adaptations; keep unknown baselines null, distinguish buyers from users and keep hybrid offer goals separate. The optional read-only scripts/marketing_library.py validates selections without executing work. Existing provenance, Phone Capture Fit Rule, channel contracts, client isolation, human approvals and independent ECO verdicts still govern every handoff.

A marketing plan the founder can start executing on Monday, matched to the company's actual stage — so that effort goes to the constraint that is currently binding growth rather than to the tactic that is currently fashionable. The plan names what to do, what to deliberately not do, which growth loop to build, and what to measure.

The wrong strategy at the right stage wastes money. Stage diagnosis is the load-bearing judgment in this skill.

Done when

Work type strategy-plan — floor E3/C3/O1 (harness/eco-floors.yaml).

  • E3 — a named human approved the exact plan document.
  • C3 — the plan clears banned_word_check, and someone other than its author read it end to end.
  • O1 — every P0 recommendation names the metric it targets, with a baseline, a threshold, and an owner. A plan whose first work item has no metric is not finished.

A plan nobody executes is not CLOSED. Its outcome is plan_adopted / first_action_shipped, read at 30 days.

Constraints

  • Stage first. Do not produce recommendations before diagnosing stage from revenue, retention, team, and current channels. A plan built on an assumed stage is worse than no plan.
  • Read MARKETING.md from the project root before asking the user anything. If it does not exist, build it from the codebase — README, manifests, landing pages, route files, analytics and email config — and confirm the draft. Do not open with discovery questions the repo can answer.
  • Provenance. Any quantitative or client-facing claim — benchmarks, CAC, conversion rates, competitor traffic, market size — runs through the collector first and cites a source. Missing data is a data gap, never a benchmark. See harness/references/audit-data-provenance.md.
  • Anti-patterns are mandatory output, not a bonus. What not to do at this stage is the half of the plan that saves money.
  • Phone Capture Fit Rule applies when the business appears phone-led. Compare alternatives, and do not lead with a phone answering service or AI receptionist when phone demand is low or the workflow is self-serve by design.
  • No spend commitments. Budget allocation is a recommendation; it does not authorize spend.

Revenue lens

Load the Revenue Lens section of harness/references/audit-data-provenance.md with the provenance rule.

  • Open with the funnel, not activity. Traffic source → intent → lead → follow-up → booked/paid, one row per stage, before any channel or technical finding. Sessions and lead counts are inputs, never the headline.
  • Rank gaps by the dollars each stage leaks. High-intent traffic with no lead path, leads with no follow-up, and follow-up with no booking outrank traffic volume.
  • Technical setup is one line. Schema, llms.txt, AI crawler access and sitemap get a sourced pass/fail; they become findings only when a measured check fails.
  • Unverified means check. Any claim or funnel stage not verified in this run is labeled check with what would verify it, is not scored, and is mirrored in _data-gaps.md.

Context

NeedLoad
Stage definitions and stage-specific movesknowledge/playbooks/marketing-by-stage.md
Which growth loop fits the productknowledge/playbooks/growth-loops-applied.md
Demand generation mechanicsknowledge/playbooks/demand-generation.md
Metrics that matter per stageknowledge/playbooks/saas-metrics-guide.md
Distribution, first growth hire, channel coverageknowledge/playbooks/growth-hacker-first-hire-os.md
Budget frameworksknowledge/playbooks/marketing-budget-forecasting.md
Which framework governs a conflictknowledge/_arbitration.md + knowledge/frameworks/marketing-science/diagnosis-first-operating-order.md
Product, ICP, voice, current channelsMARKETING.md (project root)

Stage map — the one table worth carrying inline, because it anchors every downstream call:

StageRevenueGoalMarketing mode
Pre-Launch$0Validate demandTalk to humans, build waitlist
Early$0–$10K MRRFind PMF + first channelManual, unscalable, learn what works
Growth$10K–$100K MRROptimize + expand channelsSystematize what works, test new channels
Scale$100K+ MRRSystematize + build teamHire, automate, diversify

Output goes to workspace/growth-plan/: stage assessment, 90-day plan, budget allocation, metrics dashboard, skill routing, anti-patterns. Same paths as v1 — downstream tooling does not branch on version.

Routing. The plan should end by naming which /kai skills execute it and when. This skill identifies which growth loop to build; it does not design the loop.

Escalate when

  • Revenue, retention, or channel data is unavailable and the stage cannot be diagnosed — say so rather than guessing a stage.
  • The user's stated stage conflicts with their numbers.
  • The plan would require spend the user has not authorized.
  • The business is in a regulated category where the obvious channel recommendation carries compliance risk.
  • Two loaded frameworks give conflicting guidance and knowledge/_arbitration.md does not resolve it.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall