Content
50%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 well-organized and token-efficient, but it lacks executable guidance — scripts are named without invocations or code — and its workflow has no explicit validation checkpoint. The referenced script files also do not exist in a bundle, weakening navigation.
Suggestions
Show concrete invocations for the scripts (e.g. `python calculate_ratios.py --input statements.json --ratios all`) or include a short executable code snippet so the guidance is copy-paste ready.
Insert an explicit validation checkpoint in the workflow (e.g. 'Validate data completeness; only proceed to calculation when all required line items are present').
Either bundle the referenced calculate_ratios.py and interpret_ratios.py under scripts/ and link them, or remove the dangling references so navigation matches the actual bundle.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly tight bullet lists (Capabilities, Input/Output Format, Best Practices, Limitations) with no padding about concepts Claude already knows; only the intro line 'comprehensive financial ratio analysis for evaluating company performance...' is slightly trimmable, keeping it just below the fully-lean score 5. | 4 / 5 |
Actionability | It names scripts ('calculate_ratios.py', 'interpret_ratios.py') but gives no invocation, arguments, imports, or executable code, and the 'How to Use' steps are high-level hints ('Provide financial statement data', 'Select Ratios') without the specific steps to execute — matching the 'minimal concrete guidance, missing specific steps' anchor rather than the incomplete-pseudocode score 3. | 2 / 5 |
Workflow Clarity | A rough sequence exists (Input Data → Select Ratios → Interpretation) and Best Practices mentions 'validate data completeness before calculations', but validation is a suggestion rather than an explicit checkpoint in the flow, fitting the 'steps listed but validation gaps, checkpoints missing or implicit' anchor; it does not reach a 4 because there is no concrete validate-then-proceed step. | 3 / 5 |
Progressive Disclosure | Sections are clearly organized, but the 'Scripts' section references calculate_ratios.py and interpret_ratios.py that are not present in any scripts/ bundle directory, so the references are dangling and not clearly signaled against an actual bundle — placing it at 'some structure, references present but not clearly signaled' rather than the well-placed score 4. | 3 / 5 |
Total | 12 / 20 Passed |