ワンクリックで
know-your-unknowns-skills
know-your-unknowns-skills には Infinite-Labs-AI から収集した 12 個の skills があり、リポジトリ単位の職業カバレッジとサイト内 skill 詳細ページを表示します。
このリポジトリの skills
Route and execute unknown-discovery workflows before, during, or after implementation. Use when a user asks for a blindspot pass, unknown unknowns, "what am I missing", domain vocabulary bootstrapping, design directions, mock-before-wire artifacts, codebase-grounded brainstorms, one-question-at-a-time interviews, reference semantics maps, tweakable plans, implementation notes, buy-in docs, merge quizzes, or interactive review artifacts/HTML artifacts that expose assumptions and capture human decisions.
Surface unknown unknowns before touching unfamiliar code or systems. Use when the user asks for a blindspot pass, "what am I missing", hidden traps, implementation risks, prompt hardening, or wants Codex to inspect a module/repo area before writing code.
Package completed work into a stakeholder buy-in/sign-off artifact. Use when the user asks for a pitch doc, sign-off doc, Slack-ready summary, rollout package, objection handling, or a review artifact that leads with demo/evidence and names who must approve what.
Generate multiple incompatible product/UI/design directions so the user can discover taste by reaction. Use when the user says they have no visual taste, will know it when they see it, wants 3-4 directions, wants alternatives before committing, or needs an interactive review artifact for design choices.
Teach the user enough domain vocabulary and mental models to prompt accurately. Use when the user says they do not understand a domain, lacks the right words, asks to learn their unknown unknowns, or needs practitioner vocabulary before asking Codex to do unfamiliar work.
Keep a durable implementation-notes/deviation log while building. Use when the user asks to log surprises, preserve deviations from a plan, keep implementation notes, pick conservative options and continue, or capture unknowns discovered during execution for a later attempt.
Expand a known problem into a codebase-grounded intervention space. Use when the user knows the outcome problem but not the best fix, asks to brainstorm interventions, wants options from cheapest to most ambitious, or wants Codex to search the repo for leverage points before choosing a plan.
Verify understanding of a diff or completed change before merge. Use when the user asks to be quizzed before merging, wants to ensure they understand what changed, needs a merge-readiness report, or wants an artifact that tests comprehension and points wrong answers back to the relevant explanation.
Build a throwaway mock or prototype before wiring production code. Use when UI layout, interaction, scope, placement, or flow is ambiguous and the user wants to react to fake data, a clickable HTML artifact, or a prototype before implementation.
Prove semantic understanding before porting, reimplementing, or matching a reference implementation. Use when the user points at another file/package/language/repo as the desired behavior, asks for a semantics map, or wants confirmation before Codex writes code based on a reference.
Interview the user one question at a time about ambiguous requirements, ordered by blast radius. Use when the user says "interview me", "do not assume", wants one question at a time, has ambiguous specs, or needs Codex to identify which answers would change architecture, scope, data model, UX, or rollout.
Write implementation plans sorted by the decisions the human is most likely to tweak, with mechanical work buried below. Use when the user asks for a tweakable plan, decision-first plan, plan review artifact, alternatives before sequencing, or wants judgment calls surfaced before implementation.