Launch Message House Builder
Turn an approved positioning canvas into traceable launch messaging, a working-backwards PR-FAQ spine, and channel-specific message angles.
The skill explicitly treats supplied material as untrusted, scopes writes, asks permission before saving, and routes claims and launch facts through authorized proposal events. However, runtime permission boundaries, rollback behavior, and proposal-execution safety are not fully verifiable from the supplied file, so points are deducted.
The contract defines inputs, outputs, completion criteria, dependency states, and stop conditions in useful detail. Static review provides no execution or dedicated key-path test evidence, and the referenced runtime and protocols cannot be reproduced from the selected file alone; the score is therefore capped and reduced.
Triggers, audience scenarios, required inputs, output shape, and non-fit boundaries are unusually explicit, with Chinese-language terminology included. The skill nevertheless depends on a completed positioning canvas and accepted Narrative/claims projections, while channel requirements require external verification, so points are deducted.
The file includes versioning, Apache-2.0 licensing, compatibility, Quick Starts, a contract, handoff requirements, references, and next-skill guidance. Maintenance ownership, changelog, and update path are not clear, and SECURITY.md states 17.x is supported while this skill declares 18.0.0, so points are deducted.
The declared outputs are concrete: a message house, PR-FAQ spine, channel angle packs, evidence labels, and a handoff summary. They appear potentially usable for downstream asset production, but no representative output or independent execution evidence is supplied, and completion depends on upstream artifacts; the static cap limits this score.
The skill requires evidence labels, canon/version identifiers, claims offsets, and dependency status, and points to repository protocols. There is no skill-specific test suite, third-party validation, or reproducible output; the supplied tests mainly cover shared architecture and audit validation, so points are deducted.
- Before use, confirm that the positioning canvas, Narrative canon, and claims projection are accepted and that their offsets are valid; missing inputs must not produce canonical or publish-ready messaging.
- The referenced registry-events.py, skill-contract.md, and official channel specifications are not included in the selected file and should be checked separately for availability, version alignment, and permission boundaries.
- SECURITY.md lists 17.x as the supported line while the skill declares 18.0.0; verify the current support policy and maintenance path.
- The output is an angle brief rather than finished copy, so claims adjudication, channel-spec verification, and human review remain necessary before publication.
What does this skill do, and when should you use it?
This skill builds the messaging layer for a product launch from an existing positioning canvas and accepted narrative and claims projections. It produces a tagline, one-liner, three value pillars, persona-level evidence labels, a launch-day PR-FAQ spine, and channel angle packs. It does not create positioning, finished content, ad or email units, or claim adjudications. It fits teams that need launch messaging to remain coherent, evidence-labeled, and ready for a controlled handoff to asset production.
Reads the positioning canvas, personas, channels and tier, plus named offsets in memory/projections/narrative.json and memory/projections/claims.json; verifies that the required positioning, narrative and claims inputs are usable; derives a tagline, one-liner, three value pillars and persona-specific proof points labeled Measured, User-provided or [needs source]; writes a launch-day-tense working-backwards PR-FAQ spine with five external and five internal FAQs; creates per-channel angle packs containing angles, lead proof points and target personas rather than finished copy; collects unresolved claims, dependency status and the standard handoff summary; and writes permitted outputs under memory/launch/message-house-builder/.
- A product team has completed positioning and needs a coherent message house for a launch.
- A launch lead needs a launch-day-tense PR-FAQ that addresses buyer objections and difficult internal questions.
- A marketing team needs distinct message angles for Product Hunt, press, store listings or an announcement email without producing final copy yet.
- A brand team needs persona-level proof points with explicit evidence labels instead of unsupported metrics.
What are this skill's strengths and limitations?
- Produces a defined message house, PR-FAQ spine and channel angle packs.
- Requires evidence labels for persona proof points and preserves unresolved `[needs source]` claims.
- Clearly separates positioning, messaging, finished assets and claim adjudication responsibilities.
- Includes a structured narrative and claims dependency handoff.
- Requires a usable positioning canvas plus accepted narrative and claims projections; missing inputs block canonical launch messaging.
- Produces channel angles, not finished blog posts, pages, ads or email units.
- Unresolved claims block release use and require the offer-claims-registry workflow.
- The supplied source does not provide an independent test suite or platform-coverage evidence for this skill.
How do you install this skill?
Install the full collection with npx skills add aaron-he-zhu/aaron-marketing-skills. In Claude Code, use /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. The README documents selecting a single skill with the generic host -s option; this skill is at launch/assemble/message-house-builder/SKILL.md and has slug aaron-message-house-builder.
How do you use this skill?
Prepare a completed positioning canvas, then provide the personas and launch channels and trigger: Build a message house for [product] from the positioning canvas. Personas: [list]. Launch channels: [list]. You can also ask: Write a working-backwards PR-FAQ for our [launch type] — launch-day tense, five external + five internal FAQs. If the positioning canvas is missing, route to positioning-mapper rather than treating a fallback as canonical.