Skill UI

This internal agent skill defines the structured procedure for displaying bug triage results in terminal markdown format. It generates comprehensive summaries of bug clusters, provides detailed views of evidence and reports, and handles the parsing and execution flow for user review commands (e.g., `details`, `file`, `dismiss`, `escalate`). It is a core component for automated bug triage and defect management workflows.

This skill enables deep customization of the Webiny Admin UI. Developers can implement white-labeling (custom logos, titles, and theme colors), add highly specific custom columns to content entry list views, and create bespoke forms for various page types. It allows extending core components like dialogs, drawers, and the Lexical editor, making the admin panel fully branded and feature-rich.

This skill enables executing implementation plans by assigning isolated sub-agents to each task, ensuring high quality through task-specific reviews and a final branch-wide audit. It promotes continuous execution without unnecessary human interruptions, allowing autonomous decision-making for conflicts and ambiguities while maintaining a progress ledger for recovery.

This tool assists engineers in transforming vague work logs into structured, evidence-backed impact statements for professional documentation. It uniquely mines Copilot CLI session logs, git commits, and GitHub PRs to reconstruct forgotten work. It enforces a mandatory 'Action → Result → Evidence' impact contract, making it indispensable for preparing performance reviews, promotion packets, and self-assessments.

This skill generates comprehensive, single-page engineering runbooks. It structures critical operational information, including service overviews, alert response tables, common deployment procedures, on-call rotations, and incident response checklists. It is essential for DevOps, SRE, and engineering teams creating standardized operational documentation (Ops Docs).

Logic Lens is an advanced AI skill that performs deep, formal logic-driven code review. Unlike standard linters, it analyzes code for complex issues like race conditions, security vulnerabilities (SQL injection, XSS), type safety flaws, and algorithmic errors across 9 critical risk categories. It is ideal for pre-merge PR reviews, auditing sensitive code paths, and refactoring complex business logic.

A tool that generates and opens a GitHub Pull Request (PR) dashboard in the browser. Users can specify precise criteria, including date ranges (e.g., 'last 7 days', 'this month') and specific roles (e.g., 'Authored by me', 'Requested reviews'), to view detailed PR activity. Requires GitHub CLI authentication for full functionality.

This skill provides comprehensive code reviews, helping users assess code quality, identify various potential issues (syntax, logic, security, performance), and receive actionable suggestions for improvement. It generates a detailed, structured report covering a summary, found issues, and quality metrics.

This orchestrator converts complex Markdown content—such as long-form specs, code reviews, or slide decks—into a single-file, lightly-interactive HTML document. It deterministically classifies the input type (Document, Review, or Slides) and routes the content to the appropriate specialized converter sub-skill, ensuring the output retains density and structure far beyond what raw markdown can achieve.

This skill helps developers fix social sharing previews, ensuring that URLs render as rich, consistent cards across major platforms like Facebook, LinkedIn, X/Twitter, and WhatsApp. It provides standardized implementation for Open Graph (OG) tags and Twitter cards, crucial for improving SEO and discoverability.

Comprehensive tool for finding and fixing WCAG 2.2 accessibility violations across codebases or live web pages. Operates in two modes: Report mode generates a prioritized, detailed, non-editing audit report (ideal for code reviews), while Fix mode performs an audit-edit-verify loop directly on the target to apply necessary fixes.

This meta-skill scans a target codebase path, intelligently determines the most appropriate architectural diagram type (e.g., sequence, component, class), and concurrently generates the diagram in two standardized formats: PlantUML source code and draw.io XML. Finally, it composes these deliverables into a cohesive architecture document (DOCX). Ideal for creating comprehensive onboarding guides or technical RFCs that require both text-friendly and highly editable views of the system architecture.