Content
57%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.
The body is an efficient, well-organized strategic overview with genuinely useful domain content (metric targets, red flags, decision triggers), but its effectiveness is undermined by the bundle itself: every detailed reference and every executable script it points to is missing, and the Quick Start commands lack the input/output specifications needed to run them. As shipped, the skill functions as a dashboard of CFO heuristics rather than an actionable toolkit.
Suggestions
Ship the referenced files or inline their essentials: all six bundle paths (references/financial_planning.md, references/cash_management.md, references/fundraising_playbook.md, scripts/burn_rate_calculator.py, scripts/unit_economics_analyzer.py, scripts/fundraising_model.py) are referenced but absent, breaking the skill's core navigation and executability.
Specify inputs and outputs for each Quick Start script — expected data format, CLI arguments, and what the script produces — so the commands are executable rather than bare invocations that cannot be acted on.
Add a sequenced workflow with validation checkpoints for at least the flagship deliverables (e.g., board package or financial model: gather inputs → build → cross-check totals/tie-out against actuals → deliver) instead of deferring all process detail to the missing reference files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Dense, table-driven content with no explanations of concepts Claude already knows; every section adds domain-specific value (targets, thresholds, red flags). Not 5 because of trimmable duplication: reference paths appear in both the Core Responsibilities table and the Resources section, and the Keywords section re-lists the frontmatter description's triggers. | 4 / 5 |
Actionability | Concrete thresholds and commands are present ("python scripts/burn_rate_calculator.py", "Burn Multiple < 1.5x", "CAC Payback < 18 mo"), but the three Quick Start scripts are invoked with no arguments or input-format specification, and none of the six referenced bundle files exist — the commands are not executable as shipped. This is more than the minor gaps of anchor 4, but the specific targets, red flags, and artifact specs keep it well above anchor 2's minimal guidance. | 3 / 5 |
Workflow Clarity | Entry sequencing exists ("Key Questions (ask these first)", the Output Artifacts map) and the Internal Quality Loop ("Self-verify... Peer-verify... Critic pre-screen") provides an output checkpoint, but the core workflows — model building, fundraising prep, board pack assembly — have no sequenced steps or validation checkpoints in the body; all process detail is deferred to reference files that are absent. Sequence fragments are present but checkpoints are implicit, matching anchor 3 rather than anchor 4. | 3 / 5 |
Progressive Disclosure | The body is a well-organized overview with a clearly signaled, one-level-deep Resources section — but scored against the actual bundle structure, none of the six referenced files (three references/*.md, three scripts/*.py) exist, and the body additionally depends on external ../agent-protocol/SKILL.md and company-context.md. Navigation is broken, which exceeds the "minor organization gaps" of anchor 4, but the structure is good and references are prominently signaled, not buried as in anchor 2. | 3 / 5 |
Total | 13 / 20 Passed |