taketomarket 2.2.0 → 2.3.0

package/references/playwright-mcp-setup.md

@@ -0,0 +1,164 @@
+# Playwright MCP Setup — Reference
+
+**Purpose:** Step-by-step install of the official Microsoft Playwright MCP server plus the Chrome/Edge extension bridge for takeToMarket skills (competitor scan fallback, LinkedIn/Twitter author scraping, landing/pSEO verification).
+
+**Verified:** 2026-05-18 against `@playwright/mcp@0.0.75` and "Playwright Extension" v0.2.1 in the Chrome Web Store.
+
+## What is Playwright MCP
+
+Playwright is a browser automation library. The Playwright MCP server (Model Context Protocol) wraps it so AI agents can drive a real browser. Key capabilities:
+- Visit any URL and read the rendered DOM as a structured accessibility snapshot (no vision model required).
+- Take screenshots at any viewport size.
+- Interact: click, type, scroll, hover, drag, switch tabs, press keys.
+- **Connect to the user's already-running Chrome/Edge browser** via the Playwright Extension — critical for auth-gated sites like LinkedIn, X/Twitter, and any site behind SSO/2FA. The LLM never sees your credentials; it reuses an existing logged-in tab you pick.
+
+Source of truth: <https://github.com/microsoft/playwright-mcp> and <https://playwright.dev/mcp/configuration/browser-extension>.
+
+## Prerequisites
+
+- **Node.js 18 or newer** (required by `@playwright/mcp`, enforced via `engines.node`).
+- **Chrome or Edge** (the extension bridge is Chromium-only — Firefox and WebKit work for headless mode but not for the logged-in browser bridge).
+- An MCP-aware runtime: Claude Code, Codex CLI, or Cursor.
+
+## Installation
+
+### 1. Install the MCP server
+
+The server is the npm package `@playwright/mcp`. You normally do **not** install it globally — runtimes invoke it via `npx`, which fetches and caches it on first run. To pre-warm the cache and confirm Node finds it:
+
+```bash
+node -v                                  # must print v18.x or higher
+npm view @playwright/mcp version         # should print 0.0.75 (or later)
+npx -y @playwright/mcp@0.0.75 --help     # pre-warm cache + confirm version pin resolves
+```
+
+### 2. Install the Chrome extension bridge
+
+The extension is published in the Chrome Web Store as **"Playwright Extension"** (publisher: Playwright Team / Microsoft, ID `mmlmfjhmonkocbjadbfplnigmagldckm`). Note: third-party listings sometimes call it "Playwright MCP Bridge" — the official name is just **Playwright Extension**.
+
+Install path (recommended):
+
+1. Open <https://chromewebstore.google.com/detail/playwright-extension/mmlmfjhmonkocbjadbfplnigmagldckm> in Chrome or Edge.
+2. Click **Add to Chrome** (or **Add to Edge**).
+3. Pin it to the toolbar so the connection prompt is visible.
+4. Log into the sites you plan to scrape (e.g., LinkedIn, X/Twitter) in normal tabs — the bridge reuses those sessions.
+
+Sideload path (corporate machines with Chrome Web Store blocked):
+
+1. Clone or download the latest tag from <https://github.com/microsoft/playwright-mcp>.
+2. Build the extension per the repo's `extension/` directory README, or grab the prebuilt `.zip` from the GitHub release matching your `@playwright/mcp` version.
+3. Open `chrome://extensions`, toggle **Developer mode** on, click **Load unpacked**, and select the unzipped extension directory.
+
+Either way, the extension only does anything when an MCP server is launched with `--extension` (see Step 3).
+
+### 3. Configure runtime MCP settings
+
+The flag that connects the MCP server to the extension is `--extension`. Without it, the server spawns its own headless browser and ignores your logged-in sessions.
+
+**Claude Code (recommended: CLI command):**
+
+```bash
+claude mcp add playwright -- npx -y @playwright/mcp@0.0.75 --extension
+```
+
+This writes the entry to `~/.claude.json` (or `~/.claude/settings.json` depending on Claude Code version). Equivalent manual JSON if you prefer to edit by hand:
+
+```json
+{
+  "mcpServers": {
+    "playwright": {
+      "command": "npx",
+      "args": ["-y", "@playwright/mcp@0.0.75", "--extension"]
+    }
+  }
+}
+```
+
+Drop `--extension` from `args` if you want headless mode only (faster for non-auth-gated work like landing/pSEO Lighthouse checks).
+
+**Codex CLI (`~/.codex/config.toml`):**
+
+```toml
+[mcp_servers.playwright]
+command = "npx"
+args = ["-y", "@playwright/mcp@0.0.75", "--extension"]
+```
+
+Or via Codex CLI:
+
+```bash
+codex mcp add playwright npx -- -y "@playwright/mcp@0.0.75" --extension
+```
+
+**Cursor (`~/.cursor/mcp.json` for global, or `.cursor/mcp.json` per-project):**
+
+```json
+{
+  "mcpServers": {
+    "playwright": {
+      "command": "npx",
+      "args": ["-y", "@playwright/mcp@0.0.75", "--extension"]
+    }
+  }
+}
+```
+
+Cursor also supports a one-click install deeplink from the Playwright MCP README (`cursor://anysphere.cursor-deeplink/mcp/install?...`) — use it if you prefer not to hand-edit JSON.
+
+### 4. Test connection
+
+Restart the runtime so it picks up the new MCP server, then in a takeToMarket-enabled session run:
+
+```
+/ttm-playwright-setup
+```
+
+The skill performs a sanity test: opens `https://example.com`, takes a screenshot, and reports the page title. First run with `--extension` will pop up a Playwright Extension prompt in your browser asking which tab to connect to — pick one, and the server proceeds.
+
+Manual sanity check without the skill:
+
+```bash
+npx -y @playwright/mcp@0.0.75 --extension --port 8931
+# Server prints: "Listening on http://localhost:8931/sse"
+# Then in the MCP runtime call the `browser_navigate` tool with url=https://example.com
+```
+
+## Capabilities unlocked
+
+After setup, the following skills can use Playwright:
+- `/ttm-linkedin-post` — scrape author profiles for mimic style (requires `--extension` + logged-in LinkedIn tab).
+- `/ttm-competitor-scan` — render JS-heavy competitor sites that block plain `fetch`.
+- `/ttm-landing` and `/ttm-pseo` quality gates — Lighthouse audits, mobile screenshots, visual diff (works in headless mode; `--extension` not required).
+- (Future v2.4.0) `/ttm-twitter-post`, `/ttm-linkedin-engage` — both will require `--extension`.
+
+## Useful flags
+
+| Flag | Purpose |
+|------|---------|
+| `--extension` | Connect to a running Chrome/Edge instance via the Playwright Extension. Required for auth-gated scraping. |
+| `--headless` | Run with no visible window (faster, no extension support). |
+| `--browser <chrome\|firefox\|webkit\|msedge>` | Pick the browser engine. Only `chrome` and `msedge` work with `--extension`. |
+| `--isolated` | Use a throwaway profile per session; storage state is discarded on close. |
+| `--user-data-dir <path>` | Use a persistent profile directory. Different workspaces auto-namespace via hash. |
+| `--port <n>` | Run as a standalone SSE/HTTP server instead of stdio (useful for debugging). |
+
+Full flag list: `npx -y @playwright/mcp@0.0.75 --help`.
+
+## Troubleshooting
+
+- **"No tab selected" or extension popup never appears.** Make sure the Playwright Extension is enabled in `chrome://extensions` and pinned. The MCP server must be started with `--extension`; without that flag, the extension stays idle.
+- **`Cannot find module '@playwright/mcp'` or `npx` hangs.** Node version is too old. Run `node -v` — must be `v18.x` or newer. Upgrade via `nvm install 20` and retry.
+- **MCP server doesn't show up in Claude Code.** Confirm the entry exists: `claude mcp list`. If you edited JSON directly, fully quit and relaunch Claude Code — settings are read once at startup.
+- **Codex doesn't load `~/.codex/config.toml`.** Confirm the file path with `codex mcp list`. Some Codex builds use `~/.config/codex/config.toml` on Linux.
+- **Cursor "MCP server failed to start".** Cursor needs an absolute path to `node`/`npx` on macOS if launched from the Dock. Replace `"command": "npx"` with the full output of `which npx` (e.g., `/Users/you/.nvm/versions/node/v20.10.0/bin/npx`).
+- **Logged-in scrape returns the login page anyway.** The bridge connects to the **tab you picked**, not all tabs. Pick a tab that's already authenticated on the target domain. Refresh that tab once before reconnecting if cookies feel stale.
+- **LinkedIn / X rate-limits or shows a CAPTCHA.** You're driving the user's real session — the same rate limits and bot-detection apply. Slow down requests, throttle to one profile every few seconds, and never run unattended against these sites.
+- **Headless mode works but `--extension` doesn't.** Firefox/WebKit don't support the extension bridge. Force Chrome with `--browser chrome` or `--browser msedge`.
+
+## References
+
+- Playwright MCP repo: <https://github.com/microsoft/playwright-mcp>
+- Browser extension docs: <https://playwright.dev/mcp/configuration/browser-extension>
+- Chrome Web Store listing: <https://chromewebstore.google.com/detail/playwright-extension/mmlmfjhmonkocbjadbfplnigmagldckm>
+- npm package: <https://www.npmjs.com/package/@playwright/mcp>
+- Logged-in scraping walkthrough (Debbie O'Brien, Microsoft): <https://dev.to/debs_obrien/testing-in-a-logged-in-state-with-the-playwright-mcp-browser-extension-4cmg>

package/references/positioning-check-report.md

@@ -69,7 +69,7 @@ Example: 5 assets, 3 WARN findings, 1 FAIL finding = 4 / 15 * 100 = 26.7% aggreg
 
 ## Trend Comparison Logic
 
-When a prior audit entry exists in `.marketing/DRIFT-LOG.md`:
+When a prior audit entry exists in `.taketomarket/DRIFT-LOG.md`:
 
 1. Find the most recent row in the Audit Trail table where Event = `audit`
 2. Parse the Details column to extract the prior aggregate drift percentage
@@ -89,7 +89,7 @@ If no prior audit entry exists: display "First audit -- no trend data available.
 
 For each campaign in the audit window:
 
-1. Read `.marketing/CAMPAIGNS/<slug>/DEVIATIONS.md` if it exists
+1. Read `.taketomarket/CAMPAIGNS/<slug>/DEVIATIONS.md` if it exists
 2. Filter for rows where Gate = `positioning_drift` (GATE-01 deviations)
 3. These are assets where drift was previously detected but explicitly accepted by the user
 4. Include in the "Accepted Deviations" section of the report

package/references/pseo-page-anatomy.md

@@ -0,0 +1,56 @@
+# pSEO Page Anatomy — Reference (Shared)
+
+**Purpose:** Optimal structure shared by all pSEO templates (blog, use-case, comparison, alternative). Per-template anatomy + content playbook files extend this.
+
+## Universal pSEO structure
+
+### Above the fold
+1. **Breadcrumbs** — visible + Schema.org `BreadcrumbList`.
+2. **H1** — descriptive, includes target keyword, NOT clickbait.
+3. **TL;DR / Answer block** — 2-3 sentence direct answer (this is what gets quoted by AI engines).
+4. **Author + date + read-time** — trust signals.
+
+### Body
+5. **Table of contents** — for posts >800 words. Anchor links to H2s.
+6. **Hierarchical H2/H3 sections** — one H1, multiple H2s, optional H3s.
+7. **Inline definitions** — first mention of a term: `<dfn>` wrapped or bold + definition.
+8. **Citations** — link to authoritative sources, include `cite=""` on quotes.
+9. **Visual aids** — screenshots, diagrams, tables. Each has `<figure>` + `<figcaption>`.
+
+### Conversion + related
+10. **CTA block** — mid-content, contextually relevant to the article topic.
+11. **FAQ section** — 4-8 questions with Schema.org `FAQPage` markup.
+12. **Related pages** — links to 3-5 same-template or cross-template pages.
+
+### Footer block (page-specific)
+13. **About the author** (Person Schema if used).
+14. **Updated-at** stamp.
+15. **Comment count** if engagement is part of the funnel.
+
+## AI citability signals
+
+1. **Answer-first** — TL;DR block is the most quotable; AI engines extract this.
+2. **Structured data** — Article + FAQPage + BreadcrumbList Schema.org per page.
+3. **Definitions** — `<dfn>` and structured "X is Y" sentences.
+4. **Lists, not prose dumps** — AI engines prefer parseable lists for enumerable facts.
+5. **Citations + sources** — pages with sources get cited more.
+6. **Update frequency** — pages updated more recently get cited more often.
+
+## SEO signals
+
+- `<title>` ≤ 60 chars, includes target keyword.
+- `<meta description>` 140-160 chars, written for SERP click-through.
+- One H1.
+- URL slug matches keyword (lowercase, hyphenated).
+- Internal links: minimum 3 outbound to related pSEO pages.
+- External links: minimum 1-2 authoritative outbound for "linked authority" signal.
+
+## llms.txt
+
+Site root must have `/llms.txt` listing all pSEO routes with brief descriptions. Generated by `/ttm-pseo` and updated on each new route.
+
+## Performance
+
+- LCP < 2.5s (server-rendered HTML helps; SSG even better).
+- CLS < 0.1 (reserve image dimensions, no layout-shifting ads).
+- INP < 200ms.

package/references/pseo-templates/alternative-anatomy.md

@@ -0,0 +1,31 @@
+# Alternative Anatomy
+
+For pages at `/alternatives/[product-slug]` (e.g., `/alternatives/competitor-a`). Targets users explicitly searching for alternatives.
+
+## Section structure
+1. Hero: H1 "Looking for an alternative to [Competitor]?"
+2. TL;DR: 2-3 sentences naming this product as the alternative.
+3. Why people leave [Competitor] (3-5 bullets — customer-derived if possible).
+4. Why [Product] is the alternative (3-5 pillars).
+5. Feature parity table.
+6. Pricing comparison.
+7. Migration guide (concrete steps).
+8. Customer quotes from switchers.
+9. FAQ.
+10. CTA.
+
+## URL pattern
+`/alternatives/[product-slug]`
+
+## Schema.org
+- `WebPage` schema.
+- `Product` schemas.
+- `FAQPage`.
+
+## Length
+800-1500 words.
+
+## Difference from comparison page
+- Alternative pages target intent "alternatives to X". Direct funnel.
+- Comparison pages target intent "X vs Y". Research funnel.
+- Alternative pages can be more opinionated. Comparison pages must be balanced.

package/references/pseo-templates/alternative-content-playbook.md

@@ -0,0 +1,32 @@
+# Alternative Content Playbook
+
+## Hero
+
+Direct: "Looking for an alternative to [Competitor]? [Product] is built for [specific JTBD]."
+
+## Why people leave [Competitor]
+
+3-5 reasons. Source from:
+- G2/Capterra negative reviews of competitor.
+- Reddit/HN threads.
+- ICP.md customer-language library entries that complain about competitor.
+
+Be specific. "Pricing got too expensive" is OK. "Bad product" is not.
+
+## Migration guide
+
+Step 1: Export from [Competitor].
+Step 2: Import to [Product].
+Step 3: Verify [specific things].
+Step 4: Cancel [Competitor].
+
+Include code snippets, screenshots, links to docs.
+
+## Customer quotes
+
+Switchers' words are gold. Highest-converting content on these pages.
+
+## CTAs
+
+Primary: "Migrate to [Product] free for 30 days".
+Secondary: "Read migration docs".

package/references/pseo-templates/blog-anatomy.md

@@ -0,0 +1,28 @@
+# Blog Anatomy
+
+Extends `references/pseo-page-anatomy.md`. Specific structure for blog posts under `/blog/[slug]`.
+
+## Section structure
+1. Hero with H1 (keyword-rich descriptive title).
+2. TL;DR (2-3 sentences).
+3. Author + date + tags.
+4. Intro paragraph (hook + promise + scope).
+5. Section H2s — typically 4-7 sections.
+6. Code examples or visuals where applicable.
+7. Conclusion with key takeaways (3-5 bullets).
+8. FAQ section (4-6 questions).
+9. Related posts (3-5 cards).
+
+## URL pattern
+`/blog/[slug]` — slug derived from H1 lowercased + hyphenated.
+
+## Schema.org
+- `Article` schema on page.
+- `BreadcrumbList`.
+- `FAQPage` on FAQ section.
+- `Person` for author (links to author profile page).
+
+## Length
+- Default: 1200-2000 words.
+- Short-form OK if topic warrants (e.g., changelog notes): 400-800.
+- Long-form OK for definitive guides: 3000-5000.

package/references/pseo-templates/blog-content-playbook.md

@@ -0,0 +1,36 @@
+# Blog Content Playbook
+
+How to WRITE a blog post that the AI engines will cite and humans will read.
+
+## Hook (first 50 words)
+
+Three patterns:
+1. **Surprising fact:** "X is 10x faster than Y. Here's why nobody talks about it."
+2. **Problem framing:** "Every time I tried to do X, I hit the same wall. Here's what fixed it."
+3. **Counter-intuitive claim:** "You should not use [common tool]. Use [alternative]. Here's the reasoning."
+
+Avoid: "In today's fast-paced world…" / "Have you ever wondered…" — both flagged by humanizer.
+
+## TL;DR
+
+2-3 sentences that, if extracted alone, would answer the implicit question. The AI engines will quote this verbatim.
+
+## Body
+
+Per H2 section:
+- One sentence stating the claim.
+- 1-3 paragraphs supporting it.
+- One example, code snippet, or visual.
+- Optional: counter-example or anti-pattern.
+
+## Closing
+
+- "Key takeaways:" bullet list (3-5 items).
+- One paragraph synthesizing the post's perspective.
+- CTA: link to product OR link to related post.
+
+## Banned moves
+- Generic intros ("Let's dive into…").
+- Padding sentences ("As we've seen, X is important.").
+- Em-dash overuse (max 1 per paragraph).
+- Tricolons ("X, Y, and Z" structure used more than once per post — see humanizer-patterns.md rule of three).

package/references/pseo-templates/comparison-anatomy.md

@@ -0,0 +1,29 @@
+# Comparison Anatomy
+
+For pages at `/vs/[competitor-slug]` (e.g., `/vs/competitor-a`).
+
+## Section structure
+1. Hero: H1 "[Product] vs [Competitor]" + neutral subhead + primary CTA.
+2. TL;DR: 2-3 sentences fair summary.
+3. At-a-glance comparison table (10-15 rows of feature comparisons).
+4. When [Product] wins (use cases where you're better).
+5. When [Competitor] wins (use cases where they're better — honesty wins).
+6. Pricing comparison.
+7. Migration guide (1-2 paragraphs if relevant).
+8. Customer quotes (if you have any from switchers).
+9. FAQ comparing the two.
+10. Final CTA: try [Product] free.
+
+## URL pattern
+`/vs/[competitor-slug]`
+
+## Schema.org
+- `WebPage` schema.
+- `Product` schema for both products.
+- `FAQPage` on FAQ.
+
+## Length
+800-1500 words.
+
+## Honesty principle
+Comparison pages that are ONE-SIDED don't rank or convert. Be honest about competitor strengths. Trust signal.

package/references/pseo-templates/comparison-content-playbook.md

@@ -0,0 +1,35 @@
+# Comparison Content Playbook
+
+## Hero
+
+H1: "[Product] vs [Competitor]" — keep it neutral, no superlatives.
+
+## TL;DR
+
+3 sentences:
+1. Who [Product] is best for.
+2. Who [Competitor] is best for.
+3. The single biggest differentiator.
+
+## Table
+
+Rows: pricing, features, integrations, support, ICP fit, ease of use, deploy time.
+Columns: [Product] | [Competitor].
+Use ✓ / ✗ / partial / context.
+
+## When [Product] wins
+
+3-5 bullets. Specific use cases tied to differentiator from POSITIONING.md.
+
+## When [Competitor] wins
+
+3-5 bullets. HONEST. If they have a feature you don't, say so. Trust > vanity.
+
+## FAQ
+
+Common questions: "Should I migrate?" "Can I use both?" "Is pricing comparable?"
+
+## Banned moves
+- Calling competitor "outdated" or "legacy" unless factually true with citation.
+- Loaded language.
+- Inventing competitor weaknesses.

package/references/pseo-templates/use-case-anatomy.md

@@ -0,0 +1,28 @@
+# Use-Case Anatomy
+
+Extends pseo-page-anatomy.md. For pages at `/use-cases/[slug]` (e.g., `/use-cases/onboarding-emails`, `/use-cases/api-monitoring`).
+
+## Section structure
+1. Hero: H1 ("[Product] for [Use Case]"), tagline, primary CTA, hero visual showing the use case.
+2. TL;DR: who this is for + what they'll get.
+3. The problem (1-2 paragraphs).
+4. Why typical solutions fail (1-2 paragraphs).
+5. How [Product] solves it (3-5 features, screenshots).
+6. Step-by-step walkthrough (numbered, with screenshots).
+7. Customer proof (1-2 testimonials specific to this use case).
+8. ROI / outcome (specific numbers if available).
+9. Pricing CTA + linked alternatives.
+10. FAQ (use-case-specific questions).
+11. Related use cases (3-4 cards).
+
+## URL pattern
+`/use-cases/[slug]`
+
+## Schema.org
+- `Article` or `WebPage` schema.
+- `Product` schema if pricing shown.
+- `FAQPage` on FAQ.
+- `BreadcrumbList`.
+
+## Length
+600-1200 words. Use case pages convert; they don't need to be long.

package/references/pseo-templates/use-case-content-playbook.md

@@ -0,0 +1,30 @@
+# Use-Case Content Playbook
+
+## Hero
+
+H1 pattern: "[Product] for [Use Case]" — e.g., "Linear for product roadmaps".
+Subhead: outcome promise. "Ship better roadmaps in 30 minutes per week."
+
+## Problem section
+
+Use customer language verbatim from ICP.md customer-language library if available. Engineers respond to specific pain, not abstract pain. "Your team forgets what shipped last sprint" beats "Lack of organizational visibility".
+
+## Why typical solutions fail
+
+Name the typical alternatives (1-3) and the specific failure mode of each. This is implicit comparison without trash-talking.
+
+## How [Product] solves it
+
+3-5 feature pillars. Each pillar: H3, one paragraph, one screenshot/animation.
+
+## Walkthrough
+
+Numbered list with screenshots. Engineers want to see the actual UI before signing up. 5-7 steps maximum.
+
+## Outcome / ROI
+
+Specific number from a customer if you have it. Otherwise, a frame: "[Workflow X] used to take [Y]. Now it takes [Z]."
+
+## CTAs
+
+Primary in hero + final. Secondary as inline links to /pricing or /docs.

package/skills/ttm-101/SKILL.md

@@ -0,0 +1,25 @@
+---
+name: ttm-101
+description: >
+  Marketing fundamentals for engineers and developerneurs. Engineer-analogy-heavy
+  tour: positioning, ICP, channels, funnels, AEO, lifecycle. Run once to get
+  oriented. Section-by-section runnable with /ttm-101 --section <name>.
+disable-model-invocation: true
+allowed-tools: Read AskUserQuestion
+---
+
+# /ttm-101
+
+Read and follow the workflow at `${CLAUDE_PLUGIN_ROOT}/workflows/education/ttm-101.md`.
+
+## Flags
+
+- `--section <name>` — show one section only.
+- `--all` — print all sections (default).
+
+Sections: positioning, icp, channels, funnel, aeo-vs-seo, lifecycle, invariants.
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-aeo-check/SKILL.md

@@ -1,20 +1,25 @@
 ---
 name: ttm-aeo-check
 description: >
-  Check citation status across AI engines for a query. Use to assess how
-  your content appears in AI-generated answers.
-argument-hint: "[query]"
+  [DEPRECATED v2.3.0 -> removed v2.4.0] Merged into /ttm-seo aeo.
 disable-model-invocation: true
-allowed-tools: Read Write Bash Glob Grep
+allowed-tools: Read
 ---
 
-# /ttm-aeo-check
+# /ttm-aeo-check (DEPRECATED)
 
-Read and follow the workflow at `${CLAUDE_PLUGIN_ROOT}/workflows/discipline/aeo-check.md`
+Merged into `/ttm-seo aeo` in v2.3.0.
 
-This command will:
-- Check how AI engines cite your content for the given query
-- Analyze citation quality and positioning accuracy
-- Compare against competitor citations
-- Report AEO (Answer Engine Optimization) score
-- Suggest content improvements for better AI engine visibility
+Print:
+
+```
+⚠️  /ttm-aeo-check is deprecated.
+   Use /ttm-seo aeo <query> instead. Running /ttm-seo aeo now...
+```
+
+Invoke `Skill` tool with `skill: ttm-seo`, args: `aeo` plus any user-passed args.
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-affiliate-kit/SKILL.md

@@ -17,3 +17,8 @@ This command will:
 - Include approved proof points and claims from BRAND.md
 - Package UTM-tagged links and tracking guidelines
 - Produce a complete creative kit document
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-archive/SKILL.md

@@ -11,3 +11,8 @@ allowed-tools: Read Write Bash Glob AskUserQuestion
 # /ttm-archive
 
 Read and follow the workflow at `${CLAUDE_PLUGIN_ROOT}/workflows/utility/archive.md`
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-brand-refresh/SKILL.md

@@ -17,3 +17,8 @@ This command will:
 - Deprecate expired or outdated brand claims
 - Update voice and tone guidelines if needed
 - Validate changes against POSITIONING.md invariant
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-brief/SKILL.md

@@ -12,3 +12,8 @@ allowed-tools: Read Write Bash Glob Grep AskUserQuestion
 # /ttm-brief
 
 Read and follow the workflow at `${CLAUDE_PLUGIN_ROOT}/workflows/lifecycle/brief.md`
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-competitor-scan/SKILL.md

@@ -17,3 +17,8 @@ This command will:
 - Update COMPETITORS.md with fresh intelligence
 - Flag competitive threats and opportunities
 - Inform upcoming campaign briefs with competitive context
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-deploy/SKILL.md

@@ -0,0 +1,22 @@
+---
+name: ttm-deploy
+description: >
+  Deploy your /ttm-landing site to Vercel. Auto-detects best path: git-push to
+  Vercel-connected repo (preferred), vercel CLI, or VERCEL_TOKEN env. Walks
+  you through setup if no path is configured.
+disable-model-invocation: true
+allowed-tools: Read Write Bash AskUserQuestion
+---
+
+# /ttm-deploy
+
+Read and follow the workflow at `${CLAUDE_PLUGIN_ROOT}/workflows/site/deploy.md`.
+
+## Flags
+
+- `--prod` — deploy to production. Default is preview.
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->

package/skills/ttm-discover/SKILL.md

@@ -0,0 +1,17 @@
+---
+name: ttm-discover
+description: >
+  Discover phase: perform market and audience research including SERP analysis,
+  competitor content audit, and ambient narrative mapping. Use after creating a campaign.
+disable-model-invocation: true
+allowed-tools: Read Write Bash Glob Grep AskUserQuestion WebSearch WebFetch
+---
+
+# /ttm-discover
+
+Read and follow the workflow at `${CLAUDE_PLUGIN_ROOT}/workflows/lifecycle/discover.md`.
+
+## Next steps
+
+See `${CLAUDE_PLUGIN_ROOT}/templates/next-step-footer.md`.
+<!-- next-step-footer -->