Content
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is a comprehensive knowledge base on expansion and retention systems, but it severely violates token efficiency by including extensive benchmark data, industry statistics, and detailed reference tables all inline. While the content is substantively valuable and reasonably actionable for a strategy-focused skill, it reads more like a textbook chapter than a concise skill instruction. The lack of any bundle files to offload reference material means everything competes for context window space.
Suggestions
Extract benchmark tables (NRR benchmarks, metric targets, capacity planning ratios) into a separate BENCHMARKS.md reference file and link to it from the main skill
Extract detailed frameworks (PQA scoring model, health score components, closed-lost re-engagement timelines) into separate reference files like PQA-SCORING.md, HEALTH-SCORES.md, and CLOSED-LOST.md
Remove industry statistics and explanatory context Claude already knows (e.g., 'NRR measures whether your existing customer base is growing or shrinking', '77% of the largest software companies now use consumption-based pricing')
Add explicit workflow validation steps, such as 'After building health scores, validate against last 6 months of actual churn to confirm predictive accuracy before deploying automated responses'
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | This is an extremely long skill (~500+ lines) that reads like a comprehensive textbook on customer success and retention. It includes extensive benchmark tables, industry statistics ('77% of the largest software companies...', 'Companies with consumption-based pricing see 38% faster revenue growth'), conceptual explanations Claude already knows, and multiple detailed frameworks that could be in separate reference files. Much of this is reference data that inflates token cost significantly. | 1 / 3 |
Actionability | The skill provides concrete frameworks, scoring models, email cadences, and trigger matrices that are actionable for GTM strategy work. However, since this is an instruction-only skill (no code), the guidance is more descriptive than prescriptive - it presents many options and frameworks but doesn't always give clear decision rules for which to apply. The examples section at the end helps but is brief relative to the volume of content. | 2 / 3 |
Workflow Clarity | The skill has clear sequences in some areas (onboarding milestones, renewal timeline, closed-lost re-engagement phases) but lacks validation checkpoints and feedback loops. For example, the NRR improvement decision framework gives direction but no verification steps. The health score system describes what to measure but not how to validate accuracy. The 'Before Starting' questions provide a good intake process but there's no explicit workflow for how to synthesize answers into a recommendation. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no bundle files to offload detailed reference content. The benchmark tables, scoring models, tech stack lists, and detailed cadences should be in separate reference files. The 'Related Skills' section at the end references other skills but the core content is entirely inline with no progressive structure. Content like the full PQA scoring model, closed-lost re-engagement timelines, and health score component tables would be better as linked references. | 1 / 3 |
Total | 6 / 12 Passed |