writing-specs-designs
Help users write effective specs and design documents. Use when someone is creating technical specs, feature specs, design docs, or trying to communicate product requirements to engineering and design teams.
Security Assessment
About writing-specs-designs
Guidance for writing effective technical specs and design documents, distilled from frameworks and quotes contributed by seven product leaders. The workflow starts by pinning down the fidelity level the situation calls for: low-fidelity artifacts for conceptual alignment across a team, or high-fidelity specs that give engineers detailed implementation guidance. From there, the emphasis is on prototyping over polish, identifying the handful of moving pieces in a solution, and weighing long-term consequences, since temporary shortcuts often harden into permanent product decisions. Several core principles anchor the advice. Christina Wodtke's point that drawing badly on a whiteboard invites others to correct you illustrates how low-fidelity sketches accelerate a shared vision. Ryan Singer's shaping and breadboarding techniques, using fat-marker sketches that trace button-to-outcome flows, aim for specs detailed enough that engineers, product, and design all know what to build without prescribing UI colors or spacing. Tamar Yehoshua and Noah Weiss both push toward functional, even throwaway, code prototypes with real data because a mock-up cannot convey how a product will feel. Nikita Bier's view that products live and die in the pixels, and that every tap is a miracle, underscores owning granular design for consumer, zero-to-one products. Tom Conrad warns that early shortcuts become lasting legacies, and Ravi Mehta urges PMs to learn sketching tools like Balsamiq for self-sufficiency. The skill supplies coaching questions (whether the user needs team alignment or implementation guidance, whether they could prototype instead of document, what the ten-or-fewer moving pieces are, and whether every tap provides value) and flags common mistakes such as over-specified wireframes, static mocks for complex interactions, ignoring pixel-level detail, and specs no one reads. A deeper reference collects all ten insights from the seven guests. Use it when creating feature specs, design docs, or product requirements and when communicating requirements clearly to engineering and design teams.
FAQ
How do I decide between a low-fidelity and high-fidelity document?
Low-fidelity sketches are for conceptual team alignment, while high-fidelity specs give detailed implementation guidance; the skill's first step is determining which one you actually need.
Why does it push prototypes over polished mockups?
Because, as Tamar Yehoshua and Noah Weiss argue, a static mock cannot tell you how a product will feel. Real, even messy throwaway prototypes with real data let you test the actual experience.
What is fat marker sketching?
A Ryan Singer technique of breadboarding flows with a fat marker so you focus on buttons, transitions, and outcomes rather than getting bogged down in colors or spacing.
What common spec mistakes does it flag?
Over-specified wireframes that slow collaboration, static mocks for complex interactions, ignoring pixel-level detail on consumer products, and specs that no engineer actually reads.
Whose insights inform this skill?
It draws on frameworks and quotes from seven product leaders, including Christina Wodtke, Ryan Singer, Tamar Yehoshua, Noah Weiss, Nikita Bier, Tom Conrad, and Ravi Mehta, with all ten insights collected in a references file.
Install writing-specs-designs
Quick Setup:
- Copy the skill folder to
.claude/skills/ - Claude will automatically detect and use the skill
Repository
refoundai/lenny-skills