feature-launch-playbook

---

name: feature-launch-playbook description: Turn one feature/product-update spec into a complete, launch-ready kit — changelog entry, email announcement, LinkedIn variants, X thread, in-app banner, and (for major launches) a sales-enablement one-pager — plus a tier-appropriate launch checklist. A reasoning composite: the agent authors every asset; a bundled keyless script scaffolds the folder and gates which assets to produce by launch tier. metadata: version: 1.0.1 category: content type: composite

Feature Launch Playbook

A primarily pure-reasoning composite. You (the agent) author every asset. The bundled init_launch.py scaffolds the <feature-slug>/ folder and prints the exact tier-appropriate asset list + checklist (from launch_tiers.json), so you generate only what the tier calls for.

When to use

• "We just shipped [feature] — help me launch it."
• "Write launch copy / generate the full launch kit for [update]."
• "Turn this feature spec into marketing assets."

How to run

1. Intake feature (name, description, who benefits, problem solved, headline capability), launch (tier 1/2/3, date, CTA, available visuals), and voice (brand voice + anything to avoid). Ask via grouped intake if missing.
2. Scaffold + get the tier plan:

python3 ${SKILL_DIR}/scripts/init_launch.py --feature "Advanced Filtering" --tier 1 \
  --output ${WORKSPACE}/content

Prints {slug, folder, assets_to_generate, checklist, kit_path, checklist_path} and creates the folder. Tier 1 = full kit; Tier 2 = no sales one-pager; Tier 3 = changelog only.

3. Messaging foundation — outcome-driven headline (not feature-driven), subhead (what + for whom), proof point, single CTA. See scripts/launch_tiers.json → messaging_foundation.
4. Generate each asset in assets_to_generate, using the templates in launch_tiers.json → asset_templates. For the social arms, prefer delegating to create-linkedin-content and create-x-content so posts pass through the user's voice guide; use graphics-studio / social-kit for a launch graphic.
5. Write launch-kit.md (positioning block + all tier assets) and checklist.md (owner-assignable tasks) to the scaffolded folder; post as channel attachment. Push assets onward (sendgrid/brevo email, hubspot/salesforce CRM, slack) only if asked.

python3 ${SKILL_DIR}/scripts/init_launch.py --help lists all flags.

Outputs

launch-kit.md (positioning + tier-appropriate assets) and checklist.md, under a <feature-slug>/ folder in the workspace / Agent Teams channel.

Credentials / env

• Required: none. The agent is the core engine; the scaffolding script is keyless. The full launch kit (every asset + checklist) is produced key-free — that is the default.
• Optional (all paid, all with a fallback; none gate the kit):
  • SENDGRID_API_KEY / BREVO_API_KEY — If set → actually send the email announcement. If not → deliver the announcement copy as a file/handoff for the user to send (the default).
  • APIFY_API_TOKEN — delegated voice-guide scraping (generate-voice-guide). If set → auto-scrape the corpus; if not → paste-text fallback (the default).
  • UNSPLASH_ACCESS_KEY (and generative-art keys) — delegated graphic imagery (graphics-studio). If set → real/generated imagery; if not → keyless picsum/ASCII (the default).
  • CRM (hubspot/salesforce) / Slack push — only if the user opts into that distribution.

Notes & edge cases

• Headlines must be outcome-driven, not feature-driven ("Find your best leads in seconds", not "Introducing Advanced Filtering").
• Respect the launch tier — don't generate a full kit for a Tier-3 bug fix (the script enforces this by emitting only the tier's asset list).
• Include a proof point (metric / before-after) whenever one exists; keep the CTA single.