Pressure-test spec documents with LLM reasoning before writing code — catch design flaws early.
What it does
Bugs caught before code is written are free. Bugs caught after deployment are expensive. This skill adds a pressure-testing phase before implementation: Claude reads the spec or requirements, identifies the assumptions that are most likely to be wrong, finds edge cases that aren't handled, and raises the design questions that — if answered incorrectly — will cause problems later. It's the equivalent of a senior engineer reading a ticket before sprint planning and saying "wait, what happens when...". Made by knot0-com.
Use case
Before writing code for any non-trivial feature — especially when the spec was written quickly or the requirements came from a non-technical stakeholder. Use it when you want to find the problems before they become bugs rather than after.
"Read this spec and pressure-test it — what are we not thinking about?" "What edge cases does this user story not handle?" "Find the assumptions in this design that are most likely to break." "This feature was specced in 20 minutes — what questions should we have asked?" "Review these acceptance criteria for gaps before we start the sprint."
Provide the spec, user story, or requirements document.
Claude returns a structured list of assumptions, edge cases, and unanswered questions — prioritised by likelihood of causing problems.
Resolve the open questions before writing code. The output of this skill is questions, not code.
Input
A spec document, user story, set of requirements, or feature description.
Output
A structured list of: unhandled edge cases with specific examples, implicit assumptions that might be wrong, unanswered questions that will block implementation, and design decisions that were made without being stated.
npx skillsadd knot0-com/skills/vibe-testing
Requires skills.sh CLI
Four-phase debugging methodology: root cause investigation, pattern analysis, hypothesis testing, implementation. Never fixes without finding the root cause first.
Build, navigate, and test iOS apps through simulator automation — screenshots, gestures, and UI verification.
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.