Content
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is well-sectioned and stage-aware but verbose and largely conceptual: it restates domain knowledge across three near-duplicate sections, names frameworks Claude already knows, and lacks concrete procedural detail or worked examples for the core analyses. Workflows have a sequence but no validation checkpoints, and everything is inline with no progressive disclosure.
Suggestions
Collapse the overlapping Core Expertise, Capabilities, and Knowledge Base sections into one concise capability list and remove Behavioral Traits / Quality Standards that restate general competence, cutting substantial length.
Replace named-framework bullets (Porter, Blue Ocean, Rule of 40) with a concrete worked example or template for at least one core task — e.g. a bottom-up TAM calculation with inputs, formula, and steps — to lift actionability.
Add explicit validation checkpoints to the Response Approach workflow (e.g. 'cross-check TAM against two methods; if they diverge >30%, restate assumptions') and move detailed model templates into reference files under ./references/ to enable progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is padded and redundant — Core Expertise, Capabilities, and Knowledge Base restate the same five areas, Behavioral Traits and Quality Standards restate general competence, and named frameworks (Porter's Five Forces, Blue Ocean, Rule of 40, CAC/LTV) are concepts Claude already knows, matching the 'verbose; padded with unnecessary context' anchor rather than the 'mostly efficient' anchor at 2. | 1 / 3 |
Actionability | It gives some concrete guidance (stage-aware specifics in Special Considerations, required output-format fields, quality checks) but the core analytical procedures are only named, not specified — e.g. 'TAM/SAM/SOM calculations using bottom-up and top-down methodologies' with no worked steps or template, fitting 'some concrete guidance but incomplete; missing key details' rather than copy-paste-ready at 3. | 2 / 3 |
Workflow Clarity | The 10-step 'Response Approach' provides a clear sequence, but there are no explicit validation checkpoints or fix-and-retry feedback loops embedded in the workflow, matching 'sequence present but checkpoints missing or implicit' rather than the checkpoint-driven anchor at 3. | 2 / 3 |
Progressive Disclosure | No bundle/reference files exist and all content is inline in a ~313-line monolith; sections are clearly headed (so not an undifferentiated wall), but detail that should be offloaded (framework templates, model examples) is inline, matching 'some structure but content that should be separate is inline.' | 2 / 3 |
Total | 7 / 12 Passed |