Content
75%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 a lean, well-sequenced modeling template with concrete scaffolds and a validation step, consistently one notch below excellent. The main deficits are implicit key formulas, a missing failure/recovery loop around recalculation, and a stale data-sources lead-in.
Suggestions
Add a short feedback loop after recalculation, e.g. "If recalc.py reports broken references or the accretion cell doesn't populate, fix the formula and re-run before delivery."
State the core share-count mechanics explicitly (new shares = stock-portion equity value / acquirer price; pro forma shares = acquirer shares + new shares) so the EPS bridge is fully executable.
Rewrite the "Data sources" lead-in — no passages in this body reference Kensho/Daloopa/FactSet MCPs, so the paragraph should stand alone or be trimmed.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient — input checklists, exact line-item tables, and pointed notes with no explanations of concepts Claude already knows. Minor trimming is possible: the "Data sources" lead-in ("Many passages below say 'use the S&P Kensho MCP...'" — no such passages exist in this body) is stale, and the attribution/environment meta-text adds tokens without guiding the task. This sits between "mostly efficient with some unnecessary explanation" and "lean with every token earning its place". | 4 / 5 |
Actionability | Concrete, executable guidance throughout: exact table scaffolds (Sources & Uses, Pro Forma EPS line items), specific sensitivity grids (15%-30% premium × $0-$100M synergies; cash/stock mix), and a runnable recalc command. Minor gaps remain — key formulas such as exchange ratio and new shares issued are left implicit, and the sensitivity/premium ranges are hard-coded rather than parameterized from gathered inputs. | 4 / 5 |
Workflow Clarity | A clear 7-step sequence (inputs → purchase price → sources & uses → pro forma EPS → sensitivities → breakeven → output) with a validation checkpoint ("Recalculate before delivery: python .../recalc.py"). It falls short of anchor 5 because there is no explicit error-recovery loop (e.g., what to do if recalc fails or totals don't tie) and no tie-out checks between Sources & Uses and the purchase price analysis. | 4 / 5 |
Progressive Disclosure | The ~130-line body is well-sectioned with clear headers, and conventions/scripts are cleanly delegated to the external excel-author skill rather than inlined. No bundle files exist to verify, and no content that belongs in a separate file is inlined; it is not a 5 because navigation depends on an external skill's conventions without bundled fallback material for key formulas or output formatting. | 4 / 5 |
Total | 16 / 20 Passed |