Content
78%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.
Highly actionable, well-organized content with concrete examples and per-section rules that cover the common deck types. Its main weakness is the bundle structure: both referenced files are missing from the bundle, and the detailed section spec is inlined where the template file was meant to live.
Suggestions
Add the referenced files to the bundle (references/deck-frameworks.md and templates/board-deck-template.md) or remove the citations — both paths are currently broken, so a reader following 'See references/deck-frameworks.md for Sequoia format' finds nothing.
Move the detailed 11-section specification into templates/board-deck-template.md and keep SKILL.md as a lean overview (section order, narrative framework, bad-news delivery, cadence notes), which would also improve token efficiency.
Remove the '## Keywords' section: it duplicates trigger terms already carried by the frontmatter description and consumes tokens without instructing the model on how to execute the skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and prescriptive throughout (section rules, tables, good/bad examples) and never explains concepts Claude already knows, but the "## Keywords" section lists 12 trigger phrases that duplicate the frontmatter description's job and add no executable guidance. That is minor trimmable padding, placing it between the 3 and 5 anchors, closer to lean-and-efficient. | 4 / 5 |
Actionability | For an instruction-only skill the guidance is fully executable: "3 sentences. No more.", "6-8 metrics max. Use a table." with a filled example table, a good/bad executive-summary pair, a 5-step bad-news script with example phrasings, and "High confidence $2.6M, upside to $2.9M..." forecast formatting. Specific examples cover the common cases (quarterly, monthly early-stage, fundraising) via the Cadence Notes. | 5 / 5 |
Workflow Clarity | A clear sequence exists (Quick Start invocation → 11 sections in standard order → 4-act narrative → cadence variant), and per-section constraints ("The forecast must have a confidence level", "One sentence on each variance") plus the mistakes table act as implicit checkpoints. It falls short of 5 because there is no explicit final validation/review step; deck building is not a destructive or batch operation, so the cap of 3 does not apply. | 4 / 5 |
Progressive Disclosure | The body is well-organized with clear section headers and a clearly signaled References section, but it cites "references/deck-frameworks.md" and "templates/board-deck-template.md", neither of which exists in the bundle (no references/ or templates/ directories are present). Additionally, the full 11-section deck specification is inlined in SKILL.md even though the (missing) template file is described as the fill-in home for that detail — matching the 3 anchor's "references present but content that should be separate is inline". | 3 / 5 |
Total | 16 / 20 Passed |