Launch Tier Planner
Size a launch, set its targets, and define when to stop or roll back.
The skill treats pasted plans, exports, and partner documents as untrusted input, limits registry interaction to authorized propose requests, and requires user confirmation before saving results. Data sources, write locations, rollback thresholds, and security boundaries are disclosed. Deducted 6 points because per-write confirmation, failure rollback, and minimization of sensitive analytics data are not fully specified within this skill, while proposals can still change local state.
The output contract, done criteria, scope guard, visited-set, max-depth, and ambiguity-stop rules are explicit, and the skill prohibits invented industry benchmarks. Deducted 11 points because there are no skill-specific key-path tests, reproducible examples, concrete invocation commands, or detailed abnormal-input diagnostics; static scoring is capped at 10.
Triggers, inputs, Tier 1/2/3 scenarios, non-fit boundaries, handoff skills, and partnership branching are clearly described, with compatibility declared for Claude Code and compatible hosts. Deducted 3 points because Chinese-language output behavior is not explicitly defined and mainland-China network reachability is not evidenced at this skill level.
Frontmatter, version, Apache-2.0 licensing, Quick Start, Skill Contract, Data Sources, Instructions, Next Best Skill, and cross-references provide a coherent structure. Deducted 4 points because there is no skill-specific changelog, named maintenance owner, or explicit update path; publisher identity is unverified, and dependency details rely partly on shared repository documentation.
The expected output covers tier, type, effort matrix, D0/W1/M1 KPIs, risk register, kill criteria, rollback thresholds, and timeline, making the planned artifact directly usable as a draft. Deducted 8 points because no executed output or third-party execution evidence is supplied, so correctness, completeness, and marginal benefit over manual planning cannot be verified; the static rule caps this at 7.
The skill references the RAMP framework, shared contract, launch registry, and security materials, and requires user-provided or measured baselines, estimated labels, and checkable thresholds. Deducted 5 points because evidence is mainly author-authored specifications and repository tests that do not cover this skill's key paths, with no independent reproduction or third-party corroboration; static scoring is capped at 5.
- Registry proposals and memory writes still change local state; confirm the target, content, and authority before each operation and verify post-failure state.
- D0/W1/M1 targets and kill criteria depend on the user's own trailing baselines; industry numbers should not be substituted when baselines are missing.
- Chinese output quality and mainland-China network reachability are not demonstrated; confirm availability, authorization, and data-egress risk before using overseas connectors.
- No skill-specific execution tests or real outputs are provided, so this conclusion is limited to static documentation review.
What does this skill do, and when should you use it?
Launch Tier Planner is the Research-phase launch skill in Aaron Marketing Skills’ RAMP workflow. It decides whether a launch should be a Tier 1 flagship, Tier 2 targeted push, or Tier 3 changelog-level update, then declares the launch type and calibrates the required effort. It sets D0, W1, and M1 KPI targets against the user’s own baseline and builds a risk register with owners and rollback conditions. It also provides a T-8w to T+4w timeline skeleton without choosing calendar dates.
Reads the launch scope, audience impact, novelty, revenue linkage, an available positioning canvas, launch-registry records, prior launch outcomes, and the user’s own analytics baselines. It applies the three-question tier decision, declares a new-product, feature, relaunch, or partnership launch, and produces a tier-to-channel-intensity and asset-scope matrix. Partnership plans include partner lists and co-marketing responsibility splits. The skill outputs D0/W1/M1 targets labeled Estimated or User-provided, a likelihood-by-blast-radius risk register with owners, mitigations, kill criteria and rollback thresholds, a T-8w to T+4w timeline, and a handoff summary; after confirmation, results can be saved and registry facts submitted as authorized proposals.
- A product team is deciding whether a major release deserves a full, all-channel flagship launch.
- A growth lead is sizing a feature launch for a specific audience segment and tying effort to revenue impact.
- A marketing team is planning a partnership launch and needs explicit channel ownership and shared-copy approval.
- A launch owner needs pre-launch KPI targets, risk owners, and observable rollback thresholds.
What are this skill's strengths and limitations?
- Covers sizing, launch type, effort calibration, KPIs, risks, and rollback criteria in one structured output.
- Requires projections to be tied to the user’s own trailing baseline rather than an invented industry benchmark.
- Defines additional responsibilities for partnership launches, including channel ownership and shared-copy approval.
- Has explicit scope boundaries and a named next skill for launch-window planning.
- Does not choose a launch date or launch window.
- Does not build positioning, run creator-channel campaigns, conduct the launch-day runbook, or perform the RAMP audit.
- KPI quality depends on user-provided or measured historical baselines.
- Registry-grade facts require an authorized proposal and the repository’s registry machinery.
How do you install this skill?
Install the repository with npx skills add aaron-he-zhu/aaron-marketing-skills, or clone it with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. To pull one skill on a generic host, the README gives npx skills add aaron-he-zhu/aaron-marketing-skills -s launch-tier-planner. In Claude Code, add the marketplace with /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then install it with /plugin install aaron-marketing@aaron.
How do you use this skill?
Place launch/research/launch-tier-planner/SKILL.md in an Agent Skills-compatible skills directory and invoke it with a prompt such as How big should the launch of [product / feature] be? Audience: [who is affected]. Revenue link: [direct / indirect / none]. Another supported prompt is Declare tier and type for [launch], build the risk register with kill criteria, and sketch the T-8w to T+4w timeline. Provide the launch scope, audience impact, and revenue linkage; use launch-window-planner for the actual date or window.
How does this skill compare with similar options?
It complements launch-window-planner: this skill determines the launch’s size and type, while launch-window-planner selects the date and window. Creator-led launch campaigns are routed to campaign-planner.