Menu
Use when something is broken, failing, slow, throwing, flaky, or behaving unexpectedly.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Turn a loose idea into a sequenced map of investigation tickets, then drive them to resolution one at a time.
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
Use when creating, designing, writing, implementing, auditing, reviewing, improving, or optimizing conversion-focused landing pages, including PPC, lead generation, SaaS, ecommerce offer, signup, demo, trial, download, local-service, and campaign pages.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Anti-AI-slop design skill for greenfield pages, audits, redesigns, and design extraction from URLs or screenshots. Use when the user asks to build a new app or landing page, wants to redesign something, invokes Hallmark by name, or uses audit/redesign/study.
7-phase High-Level Design framework: clarify requirements, estimate scale, design components, model data, define APIs, deep dive on bottlenecks, and enumerate failure modes. Use for any system design question or production architecture task.
| name | diagnose |
| description | Use when something is broken, failing, slow, throwing, flaky, or behaving unexpectedly. |
Build a tight failing loop and find root cause before proposing fixes.
Diagnose owns investigation until there is a root cause and a red-capable loop.
Do not jump to build while the failure is still a theory.
build for the fix.diagnosing-bugs for hard bugs, performance regressions, flakiness, and
feedback-loop construction.systematic-debugging for strict root-cause discipline and multi-layer
evidence gathering.firecrawl when current external docs, issue trackers, or release notes may
explain the failure.vercel-react-best-practices when the failure is React/Next performance or
data-flow related.Done when there is one command or reliable loop that demonstrates the bug, a clear root-cause hypothesis supported by evidence, and a path to a regression test or documented reason no correct seam exists.