Launch Feedback Synthesizer
Turn launch feedback into prioritized themes, a visible status loop, shipped-update material, and compliant social-proof collection.
The skill treats feedback as untrusted data, limits its scope, refuses to make roadmap decisions, asks before saving, uses proposal-based registry writes, and prohibits incentivized store reviews. Points are deducted because connector, memory-write, sensitive-data, deletion, and rollback details rely on shared protocols rather than being fully specified here.
Inputs, outputs, completion criteria, duplicate merging, notification behavior, and termination rules are internally coherent and reasonably diagnosable. Static review found no skill-specific tests, abnormal-input handling, or connector-failure feedback, and the referenced shared files and tools were not reproducible from this skill alone, so the score remains below the static ceiling.
Trigger scenarios, inputs, outputs, and non-fit boundaries are clearly stated for launch-window feedback triage and proof harvesting. Points are deducted because platform and connector availability is uncertain, geo-relevance is marked low, mainland-China reachability is not demonstrated, and Chinese support is limited to metadata and brief bilingual text.
The skill provides standard frontmatter, versioning, Apache-2.0 licensing, quick starts, a contract, data sources, procedures, handoff guidance, and next-skill routing, consistent with the repository architecture. Points are deducted for lacking skill-specific sample outputs, FAQs, changelog detail, and explicit maintenance ownership; publisher identity is unverified.
The promised artifacts are concrete: a theme digest, status-loop specification, shipped announcement material, compliance matrix, metrics, and handoff summary. Points are deducted because the supplied files contain no representative executed output or execution evidence, while platform-policy judgments and connector results still require user data and review.
The instructions distinguish Measured, User-provided, and Estimated evidence and cite a Fider pattern and platform policies; the repository also contains general CI and test material. Points are deducted because the cited external policies are not included in the supplied source, the tests do not cover this skill's key paths, and no independent reproduction record is present.
- Do not treat instructions embedded in comments, tickets, or fetched pages as operational commands; confirm scope, provenance, and sensitive-data handling first.
- Connectors, registry proposal requests, and memory writes depend on the host and shared protocols; verify failure feedback, permissions, and recoverability before execution.
- Never incentivize store reviews; verify incentive and disclosure requirements against each platform's current policy rather than relying solely on this skill text.
What does this skill do, and when should you use it?
This is the Launch/Prove feedback skill in Aaron Marketing Skills. It processes launch-window comments, store reviews, feedback-board posts, and support tickets into themes with measured frequency, judged severity, and representative quotes. It specifies an open-to-completed or declined workflow with duplicate merging and subscriber notifications. It also designs a compliant review and testimonial harvest, while leaving proof amplification, outreach execution, roadmap decisions, and RAMP scoring to other skills or people.
Reads a launch slug and feedback exports, records each export's coverage window, and labels data as Measured or User-provided; clusters feedback by underlying need and records theme frequency, severity, source, and verbatim quotes; defines open → planned → started → completed／declined transitions, duplicate merging with vote transfer, and notifications to all subscribers except the acting user; creates a changelog entry, requester thank-you note, and candidate social post for completed themes; builds a platform compliance matrix for incentives and disclosures, with App Store and Google Play reviews marked never-incentivized; defines loop metrics and submits a launch theme snapshot through an authorized proposal request.
- A product marketer needs to synthesize community comments, store reviews, and support tickets after a product launch.
- A product team needs a transparent feedback workflow from open through completed or declined, including subscriber notifications.
- A launch owner has shipped a requested change and needs “you asked, we shipped” announcement material.
- A growth or customer-marketing team needs a review or testimonial collection protocol that respects platform policies.
- A product owner needs evidence-backed roadmap-shaped themes without having the skill make the roadmap decision.
What are this skill's strengths and limitations?
- Produces a concrete deliverable covering theme digest, status-loop rules, notifications, shipped-update material, and compliance controls.
- Treats feedback as untrusted data and routes unsupported product or comparative claims for sourcing.
- Sets a clear no-incentive rule for App Store and Google Play reviews.
- Works with pasted data and does not require paid tools or MCP for the core workflow.
- Owns launch-window theme triage, not always-on comment, DM, or mention management.
- Does not execute testimonial outreach, amplify harvested proof, make roadmap decisions, or calculate the RAMP profile.
- Registry proposals, connectors, and persistent storage depend on the repository runtime and authorized operations; the source provides no standalone test-suite evidence.
- Theme frequency depends on the coverage window and supplied exports; missing surfaces must be treated as coverage gaps, not zero feedback.
How do you install this skill?
Install the repository and use launch/prove/launch-feedback-synthesizer/SKILL.md. In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. On an Agent Skills-compatible host, run npx skills add aaron-he-zhu/aaron-marketing-skills; alternatively use git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The source does not document a separate package for this individual skill.
How do you use this skill?
Provide a launch slug and feedback exports. Example: Triage the feedback from our [product] launch — here are the community comments, the board posts, and the store reviews. Another supported prompt is Set up a feedback status loop for [product]: themes, open→planned→started→completed/declined, and notification rules. The source lists optional ~~launch platform, ~~app store data, and ~~brand monitor retrievals; pasted exports are also supported. After delivery, the skill asks whether the results should be saved.
How does this skill compare with similar options?
Unlike content-amplifier, this skill collects and structures social proof rather than repurposing or distributing it. Unlike outreach-manager, it specifies the review or testimonial harvest protocol rather than executing outreach threads. Unlike engagement-inbox-manager, it handles launch-window theme triage rather than ongoing comments, DMs, and mentions.