GitHub
postmortem-and-learning-extractor
Extracts reusable lessons after incidents, failures, or project milestones. Use to update skills and prevent future mistakes.
hohai99 AI Skills
GitHub ↗Explorez les skills IA publiés de hohai99 et consultez leurs instructions source.
Statistiques de l’auteur
Skills indexés
36Étoiles des dépôts des skills indexés
0Forks du dépôt
36Moyenne des étoiles par skill indexé
0Les étoiles d’un dépôt peuvent se répéter entre ses skills.
Catégories populaires
Indexation récente
Explorer
GitHub
regression-and-parity-check
Validates that changes maintain parity with previous system behavior and don't introduce regressions. Use after implementation or migration.
GitHub
legacy-behavior-snapshot
Captures existing system behavior before migration to ensure parity. Use at the start of any migration project.
GitHub
orchestrated-development-controller
Orchestrates the correct skills automatically across project creation and migration workflows. Use as the default controller.
GitHub
system-context-extractor
Extracts architectural, business, and operational context from existing systems. Use to build mental models of unfamiliar codebases.
GitHub
user-intent-pattern-analyzer
Analyzes user prompt behavior over time to infer stable intent patterns and preferences. Use continuously to minimize friction.
GitHub
lifecycle-health-monitor
Monitors overall project health across time, specifically looking for spec drift and debt. Use during regular project reviews.
GitHub
workspace-domain-glossary
Builds and enforces a shared domain vocabulary for the project to prevent terminology drift and confusion. Use when specs or code involve domain terms.
GitHub
mvp-scope-guard
Prevents MVP scope creep based on this project's constraints. Use during early design.
GitHub
no-guess-implementation
Implements tasks strictly according to specification without invented behavior. Use during coding.
GitHub
architecture-consistency-enforcer
Ensures new designs align with existing project architecture patterns to prevent drift and technical debt. Use when proposing or reviewing architecture.
GitHub
behavior-first-analysis
Defines observable system behavior independent of implementation details. Use before refactoring, migrating, or rewriting legacy systems.
GitHub
spec-driven-planning
Converts ambiguous requirements into strict, implementable specifications. Use before any coding.
GitHub
implementation-boundary-guard
Prevents implementers from crossing spec or task boundaries. Use during coding.
GitHub
backward-compatibility-guardian
Ensures that system changes don't break existing API consumers or persistent data structures. Use during feature modifications.
GitHub
long-term-debt-tracker
Tracks, classifies, and prioritizes technical debt (code, spec, and process) over time. Use to prevent silent rot.
GitHub
skill-evolution-engine
Proposes creation, modification, or retirement of skills based on recurring user needs and system friction.
GitHub
decision-freeze-governor
Determines whether documentation is sufficient to proceed without human clarification. Use before implementation to prevent rework.
GitHub
project-regression-checklist
Applies project-specific regression criteria to ensure existing functionality remains intact. Use during testing and review.
GitHub
self-healing-debugger
Fixes implementation bugs detected by tests while preserving spec intent. Use only after test failures classified as implementation bugs.
GitHub
self-test-generator
Generates tests directly from specifications and tasks. Use immediately after implementing a task.
GitHub
autonomous-test-runner
Executes generated tests and interprets results without human input. Use after tests are generated.
GitHub
spec-violation-detector
Detects mismatches between implementation and specification. Use when tests fail but behavior seems intentional.
GitHub
security-and-compliance-baseline
Ensures changes respect security, privacy, and compliance constraints. Use before release.
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.