Content
65%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 well-organized, token-efficient overview with genuinely concrete guidance (thresholds, artifact mappings, decision procedures). Its main weakness is the bundle, not the prose: every referenced deep-dive file and script is missing, leaving flagship workflows without steps and the Quick Start commands unrunnable.
Suggestions
Ship the referenced bundle files (references/executive_decision_framework.md, references/board_governance_investor_relations.md, references/leadership_organizational_culture.md, scripts/strategy_analyzer.py, scripts/financial_scenario_analyzer.py) or remove the dangling references so Quick Start and the Resources section resolve.
Inline a minimal step sequence for the flagship workflows (board-prep, fundraising readiness) since their detailed procedures currently live only in missing reference files.
Delete the 'Keywords' section — it duplicates the frontmatter description's trigger terms and spends context tokens without adding runtime guidance.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence (tables, stage-adaptive horizons, metric targets, no basic-concept explanations). Not 5: the 'Keywords' section duplicates frontmatter trigger terms at pure token cost, and aphorisms like 'Culture is what people do when you're not in the room' are rhetorical padding. | 4 / 5 |
Actionability | Concrete guidance throughout: runnable Quick Start commands, numeric thresholds ('> 12 months' runway, '< 2x' burn multiple), a request-to-artifact table, and an explicit Tree of Thought procedure. Not 5: the Quick Start commands point to scripts (strategy_analyzer.py, financial_scenario_analyzer.py) that are absent from the bundle, so they are not actually executable here. | 4 / 5 |
Workflow Clarity | The Communication section provides an explicit output sequence and verification checkpoints (self-verify, peer-verify, critic pre-screen, 🟢🟡🔴 tagging), but the flagship workflows (board-prep, fundraising readiness, Go/No-Go decisions) have no in-body steps — they are delegated entirely to reference files that do not exist. Not 4 because the main multi-step processes lack any reachable sequence; not 2 because the output/quality-loop sequence that is present is clear. | 3 / 5 |
Progressive Disclosure | The in-file design is good (one-level-deep, well-signaled 'See references/...' pointers and a Resources section), but scored against the actual bundle: no references/, scripts/, or assets/ directories exist, so every referenced path (three references/*.md, two scripts/*.py, culture-architect/, board-deck-builder/, ../agent-protocol/SKILL.md) is dangling and the disclosure layer cannot function. Not 4-5 because half the promised detailed content is missing from the bundle; not 2 because the SKILL.md itself is well structured. | 3 / 5 |
Total | 14 / 20 Passed |