Use when deciding how to represent a design to test or communicate it — match fidelity to the question.
Use during the runtime adversarial pass — attack the build the way real users (and bad actors) actually behave: malformed input, concurrency, network failure, refresh/back,…
Use whenever something is broken, flaky, or behaving unexpectedly — find the root cause systematically (reproduce, isolate, hypothesize, verify) and fix the cause, not the…
Use on every surface before it ships — the pre-merge craft pass. Work the polish checklist (tabular nums, optical alignment, concentric radius, shadow-vs-border, focus, type…
Use when scoping any user-facing surface — require a whole-surface design up front; ship in slices, but each slice lands in its final place, nothing bolted on to be re-placed…
Use before any change touches prod — enumerate what's stateful, what's a one-way door, what's reversible, and who's downstream, so rigor matches risk and the irreversible parts…
Use on every surface — accessibility is designed in, not bolted on. ~80% of accessibility is design decisions (contrast, hierarchy, target size, focus order, labels) made before…
Use when scoping how to land a change — prefer small, short-lived, independently-safe increments over a big-bang merge; keep the codebase deployable; gate risky changes behind…
Use when protecting working behavior across changes — keep a fast, risk-prioritized regression gate on the highest-value journeys, and quarantine-and-own a flaky test instead of…
Use to land your review as a durable artifact — the attack surface reviewed, findings at file:line (or "none"), the P-grade and minimal fix, and a clear GO / NO-GO — including on…
Use while writing any diff — make it look like the rest of the codebase wrote it: match conventions, reuse shared components, stay DRY, write for the next human, and never leave a…
Use when designing or fixing how the system is configured (env vars, json/yaml/toml files, flags, per-project/per-box settings, layered defaults) — give config a versioned schema,…
Use to keep problems distinct from solutions — map how people actually think about getting their job done (their cognition, reactions, guiding principles), deliberately separated…
Use when AI coding agents will read/write this codebase, or an AI/LLM feature is proposed — design for the AI reader, defend against AI-accelerated drift, and treat…
Use whenever a load-bearing call is made (or asked again) — record problem, options, what was chosen, why, and what was rejected, so settled questions aren't re-litigated and "why…
Use when cloud spend is rising, unexplained, or unattributed, or before provisioning expensive resources — treat cost as an engineering metric: make it visible, attribute it, find…
Use when the people who sell or talk to customers (sales, founders, support, partners) need the message in usable form — pitch deck, one-pager, demo script, objection handling,…
Use when reviewing a change, a design, or a system's readiness — ask "if this gets attacked or exploited in prod, would we even know?" Review the audit trail, the…
Use when applying AI to selling — conversation/deal intelligence, AI-assisted research/prioritization, AI SDRs, AI-aware forecasting.
Use before attacking a build that has any lifecycle, phases, or persisted progress — enumerate the states, transitions, and guards, then probe every transition under crash / kill…
Use after any incident (or near-miss) — write a blameless postmortem that finds the systemic/contributing causes, not a person to blame, and produces tracked action items so the…
Use when reviewing an LLM feature, AI agent, RAG system, or tool/MCP integration — treat the OWASP LLM Top 10 as its own attack class: prompt injection, excessive agency, tool…
Use BEFORE code is written — when a story, spec, or design is being shaped — to prevent the bug instead of catching it later: question the story, surface the missing examples and…
Use when output, status, notifications, or system messages are scattered across multiple places (panes, channels, logs, replies, toasts, emails) — consolidate them into one…
Use the moment you think you've found a bug — reproduce it yourself with exact steps (never theorize from reading code), make an intermittent failure deterministic, and grade its…
Use when a buyer pushes back (too expensive / no time / not now / need to check with X) — clarify and confirm the REAL concern before answering; separate a genuine blocker from a…
Use BEFORE picking which questions matter — reweight the load-bearing themes by the company's category and stage.
Use for any release that needs to reach users — a new product, a major feature, or a steady stream of small features.
Use when investigating a build or a suspicious area where a fixed script won't find the unknown — run time-boxed exploratory sessions guided by a charter (a mission), learning,…
Evidence group skill. Runs direct end-user interviews to saturation. Gates on access — halts if discovery-context says user_access is none or proxy-only.
Use when multiple teams/engineers each reinvent deploys/infra badly, or when building shared platform capability — make the safe, observable, compliant way the EASY way (a paved…
Use when reviewing a proposed architecture, a tech plan, or a PR for soundness — drive it to correctness/simplicity/fit, enable rather than gatekeep, flag the open gate without…
Use to validate that a design actually works for real people — when a flow is built or prototyped and you need to know if users can do the job.
Use whenever you (or AI) produce or review a UI draft — your prime reflex against generic, templated design.
Use when judging whether "good" is good enough now — 10x is the floor, the visible playbook is a commodity, and the best companies are icebergs.
Use whenever copy is drifting into a feature list, spec dump, or internal jargon — onboarding, a feature page, a release note, a tagline.
Use at the acceptance gate — verify the build by doing the user's flow yourself end-to-end and confirming persisted state, not by checking that "the parts exist per spec" or that…
Use when deciding how much process a piece of work needs — match ceremony to stakes; full rigor for one-way doors, a lean path for a contained fix.
Use when designing any interface where AI acts, generates, or decides on the user's behalf (agents, generative UI, AI suggestions, automation).
Use when reviewing anything that authenticates users, checks permissions, or exposes objects by id — broken access control is the #1 web risk; verify authz is enforced server-side…
Use when naming a product, feature, category, or pricing tier — or when an existing name confuses, over-promises, or fights the positioning.
Use when designing motion, feedback, or any single interactive moment (a toggle, a like, a save, a panel opening).
Use FIRST, before hunting any bug — write down the system's trust model (who can reach what, which callers are already trusted, what the deployment model is).
Use to validate a feature against real users — build a persona dossier and walk that specific person through the journey to find where it breaks for THEM.
Use when a QUALIFIED deal stalls — "let me think it over," goes quiet, keeps asking for more info, won't commit.
Use when assessing the ask and the unit economics — is the raise rational and what does it buy, is the revenue real and durable, and does the spend displace a budget or is it…
Use when generating or accepting AI/LLM-written code — treat the model as a fast junior, not an oracle: give it context, then verify every line; never ship plausible-but-wrong…
Use when writing any words inside the UI — button verbs, labels, empty/error/success messages, tooltips, confirmations, onboarding.
Use when the long-term vision is big but the entry point is fuzzy — find the credible first wedge that expands.
Use for the pitch, the launch story, the homepage hero, the keynote, the fundraising or vision story — any artifact that must move someone emotionally, not just inform.
Use whenever you reach for a value (color, spacing, radius, type, motion) or build a component — extend tokens and shared components in place, never fork into one-off className…
Use when you actually draft or edit the artifact — a landing page, hero, email, ad, onboarding screen, empty state, changelog, CTA — and need it to be clear, scannable, and…
Use the moment observed ≠ expected and especially when a result seems impossible (a deleted thing still appears, a fix has no effect, state lingers after teardown) — distrust your…
Use before scaling any message, after a launch underperforms, or whenever positioning is still a hypothesis (pre-users).
Use when auditing code that takes user or network input — hunt the OWASP web vuln classes by tracing untrusted input to a dangerous sink, and substantiate every finding input →…
Use when assessing the founder's ceiling — can they paint the huge future AND walk the exact next step, and do they bend reality or wait for permission.
Use when gauging how urgent the pain is, what the value is worth, or how to price/package — probe willingness-to-pay directly and quantify the cost of doing nothing.
Use when adding monitoring/alerting or when alerts are noisy/missing — instrument golden signals, make the system debuggable (metrics/logs/traces), and page humans only on…
Use when a design choice has to be justified to a founder/non-technical stakeholder, when proposing to spend (or take on) time/money/debt, or when "why does this matter to the…
Use when asked "what should we build (next / at all)?", when the roadmap is turning into a stakeholder wishlist, or for a build-vs-buy call — set the strategic bet first and…