Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a useful conceptual framework for product thinking with four distinct modes, each with structured question sets. Its main weaknesses are the lack of concrete output templates (e.g., a PRODUCT-BRIEF.md skeleton), missing validation/feedback loops between steps, and some unnecessary conversational filler. It reads more as a thinking framework than an actionable skill with executable guidance.
Suggestions
Add a concrete PRODUCT-BRIEF.md template or schema showing the expected output structure, so Claude knows exactly what artifact to produce.
Include decision criteria or a flowchart for selecting which mode to use based on user input signals.
Add validation checkpoints within modes — e.g., in Mode 1, after gathering answers to questions 1-3, verify the problem is real before proceeding to solution framing.
Remove conversational filler like 'Like YC office hours but automated' and 'not more founder-theater' to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably structured but includes some unnecessary filler phrases ('Like YC office hours but automated', 'not more founder-theater', 'not vibes') and could be tightened. Some explanatory text is redundant given Claude's intelligence, but it's not egregiously verbose. | 2 / 3 |
Actionability | The skill provides structured question frameworks and scoring rubrics (ICE scoring, PMF signals), which give concrete guidance. However, the instructions are more like checklists of what to think about rather than executable steps — there are no concrete commands, file templates, or copy-paste-ready artifacts. The 'output a PRODUCT-BRIEF.md' instruction lacks a template or schema. | 2 / 3 |
Workflow Clarity | Each mode has a numbered sequence of steps which provides decent structure. However, there are no validation checkpoints, no feedback loops for when diagnostics reveal ambiguity, and no clear decision criteria for choosing between modes. The handoff to 'product-capability' is mentioned but the trigger conditions are vague. | 2 / 3 |
Progressive Disclosure | The content is organized into clear modes with an integration section pointing to related skills. However, all four modes are fully inline rather than being split into separate reference files, making this a moderately long single document. The cross-references to other skills/lanes are helpful but the internal structure could benefit from splitting detailed mode instructions into separate files. | 2 / 3 |
Total | 8 / 12 Passed |