Content
86%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A lean, highly actionable interrogation skill: exact thresholds, runnable commands, a concrete output template, and clean single-file organization. Its two weaker spots are unspecified input-file provenance for the referenced scripts and a workflow lacking explicit validation or feedback steps.
Suggestions
Specify where the script input files (cohorts.json, customers.json, book.json) come from — e.g., a one-line note on their expected schema or the command/source that generates them — to close the actionability gap.
Use stable paths for the three analyzer scripts (or reference them via the Related skill link) rather than fragile ../../../c-level-advisor/... relative traversal, so the Workflow commands remain executable regardless of where the skill is installed.
Add a brief validation checkpoint to the Workflow (e.g., what to do if a script errors or if GRR data is missing) to give the sequence an explicit feedback loop.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is threshold-dense (GRR >= 90% at growth stage, 70/30 base/variable comp, $300K-$1M ARR/CSM tiers) with zero explanation of concepts Claude already knows; even the persona one-liners ('Not NRR. Gross.') encode decision rules, so every token earns its place. | 5 / 5 |
Actionability | Copy-paste-ready script invocations, exact numeric cutoffs, and a fill-in output template make the guidance highly actionable, but the input files (cohorts.json, customers.json, book.json) have unspecified provenance and the cross-skill relative paths (../../../c-level-advisor/...) are fragile, leaving minor gaps. | 4 / 5 |
Workflow Clarity | The Workflow section gives a clearly ordered 3-step sequence with an explicit start ('always start here') and a conditional gate ('if making CS team changes'), and the questions embed interpretation checkpoints ('If GRR < 85% but NRR > 100%... expansion is masking'), but there are no explicit validation or error-recovery steps. | 4 / 5 |
Progressive Disclosure | A single-file skill cleanly sectioned (When to Run, Six Questions, Workflow, Output Format, Routing, Related) with clearly signaled one-level-deep external pointers (companion skill, agent file) and no bulk content inlined that belongs in a separate file. | 5 / 5 |
Total | 18 / 20 Passed |