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 concise, well-structured single-purpose skill body with a clear workflow and validation checkpoint; the only meaningful gap is the lack of concrete nav MCP tool invocations, which caps actionability at 4.
Suggestions
Name the specific nav MCP tools (and their key parameters) to call when pulling commitment %, P&L components, fee/expense totals, and waterfall outputs, so the recompute step is copy-paste executable.
Add an explicit re-run/escalation loop for flagged mismatches (e.g., re-pull the disputed input and re-compare before promoting a flag) to strengthen the workflow's feedback loop.
Optionally state what to do when a 'next period's draft' is unavailable so the ending/beginning-capital rollover check has a defined fallback.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and efficient throughout: a formula block, a tolerance value, and brief directives with no padding or explanation of concepts Claude already knows, so every token earns its place per the 5 anchor. | 5 / 5 |
Actionability | Gives a concrete recomputation formula, an explicit 0.01 tolerance, and a worked flag example, but stops short of naming the actual nav MCP tool calls to invoke, leaving a minor execution gap versus the copy-paste-ready 5 anchor. | 4 / 5 |
Workflow Clarity | Clear sequence (Recompute → Compare → Additional checks → Output) with an explicit comparison/tolerance checkpoint; the operation is non-destructive ('Do not edit the statement') so the destructive-cap does not apply, but there is no explicit re-run/escalation loop on flagged mismatches, keeping it at 4. | 4 / 5 |
Progressive Disclosure | Under 50 lines with no bundle files and no need for external references, yet organized into clear, well-labeled sections, qualifying for the simple-skill exception at the 5 anchor. | 5 / 5 |
Total | 18 / 20 Passed |