Scan to join WeChat groupBrowse curated skills with source links, package snapshots, README assets and install signals in one calm, searchable catalog.
ComfyUI node execution lifecycle - caching, fingerprint_inputs/IS_CHANGED, validate_inputs/VALIDATE_INPUTS, check_lazy_status, execution order. Use when debugging execution, implementing caching contr…
Risk-based testing standards with structured test design. Use when adding tests, evaluating coverage, or validating edge cases. Do NOT use for test execution during do-execute — that skill has its own…
Perform multi-perspective reviews using the Discourse pattern. Handle both Tech Spec reviews and Epic integration code reviews, and separate consensus/non-consensus according to the dispute resolution…
Verify plan consistency, generate summaries, maintain plan health. Use for review, verification, summaries, or plan maintenance. Mostly read-only with maintenance operations.
Pre-implementation validation framework requiring ≥90% confidence before coding. Prevents wrong-direction work by assessing duplicates, architecture alignment, documentation, OSS references, and root …
When the user wants to implement or optimize cycle counting programs, improve inventory accuracy, or design physical inventory processes. Also use when the user mentions "inventory accuracy," "physica…
PostgreSQL performance optimization and best practices. Use when writing SQL queries, designing schemas, implementing indexes, or debugging database performance. Covers query optimization, connection …
Use when creating or extending WebdriverIO test structure from a markdown test plan by adding describe/it/hooks and pseudo-code comments without implementing WDIO commands yet.
Unit testing patterns, TDD workflow, test factories, fixtures, pure function testing, class testing. Activates when user mentions "unit test", "TDD", "test-driven", "factory pattern", "test fixtures",…
Analyze the metric data in sprint-status.yaml to suggest model allocation optimization and difficulty retagging. It is a read-only skill that outputs results only through conversation without changing…
Reflect the recurring patterns found during code reviews in the sprint into conventions.md. Accumulate the rule sources of Convention Guard to improve the quality of reviews in the next sprint.
After grasping the project context, conduct a code review from an expert's perspective and, after user approval, perform parallel improvement work through the Agent Team SPAWN. Triggers: 코드리뷰, 코드 리뷰, …
A comprehensive guide to implementing Clean Architecture in software projects, based on the boot-backend-go-clean template.
When the user wants to solve lot-sizing problems with time-varying costs or demand, handle non-stationary inventory systems, or optimize replenishment with changing parameters over time. Also use when…
Execute implementation plans using fresh subagents per task with two-stage review (spec compliance + code quality). Use when executing multi-task plans with independent work units. Enhances ralph-loop…
Use when investigating a failing WebdriverIO test and you need to understand what the test was doing and where it broke.
Migrate Keboola Python packages and components to modern uv build system with deterministic dependencies and ruff linting.
Use the Monitor pattern to coordinate the implementation of all Stories within an Epic. Delegate all difficulty level Stories to the agent, and only the Lead updates the sprint-status.yaml.
Produces structured code quality assessment reports for Frontend (Mobile & Web) or Backend codebases (IEEE 730/1016/829, Clean Code, OWASP). Use when the user asks for a code quality report, quality a…
Fetch devtap output for the current turn. Default to concise summaries and print raw logs only on explicit request.
When the user wants to optimize airline cargo operations, manage air freight, or improve cargo revenue. Also use when the user mentions "air cargo optimization," "cargo capacity management," "freight …
When the user wants to optimize hospital supply chain operations, manage medical materials, optimize PAR levels, or improve hospital inventory management. Also use when the user mentions "hospital sup…
Use when investigating and documenting website structure, components, functionality importance, and version-to-version differences for human understanding and test planning.
Use when diagnosing and coordinating fixes for failing WebdriverIO tests across context gathering, test runs, and implementation skills, while limiting retry loops and avoiding rabbit holes.