llm-skills

Generates syntax-valid, render-safe Mermaid.js diagrams (flowcharts, sequence diagrams, state diagrams, ERDs, Gantt charts). Use when a user asks to draw, visualize, map out, or document systems, workflows, API sequences, database schemas, or architectures. Don't use for generic markdown tables, text-based descriptions, or non-visual data serialization.

Converts vague feature requests into structured product requirements (PRDs) and sprint-ready user stories. Use when scoping and planning a feature in a Product Manager or Scrum Master capacity. Don't use to write or modify application code.

Standardizes how agents update and manage README.md files, adopting the Deep Insight/Strands SDK documentation style. Use when creating a new README from scratch, improving an existing README, or converting technical docs to user-friendly formats. When creating or editing Mermaid diagrams in the README, load and use the `generating-mermaid-diagrams` skill. Don't use for API-only documentation or internal technical specs.

Orchestrates specialized subagents in parallel to keep the main thread idle and maximize concurrency. Use when a task can be decomposed into independent subtasks (such as concurrent code changes, parallel research, or separate validation steps), when a task is long-running and the parent thread should remain idle while waiting, or when interactive input is needed to refine features. Don't use for simple sequential tasks, single-file edits that cannot be parallelized, or when subagent support is not available in the workspace.

Provides guidelines, visual specs, and accessibility rules for Google-branded marketing websites. Use when building or modifying web pages, landing pages, or components (such as buttons, forms, carousels, or chips) to ensure they conform to Google's layout, design, accessibility, and legal standards. Don't use for mobile native app development, generic backend service design, or premium creative visual styling (micro-animations, artistic designs)—use `designing-tasteful-frontends` for that.

Use this skill to review code. It supports both local changes (staged or working tree) and remote Pull Requests (by ID or URL). It focuses on correctness, maintainability, and adherence to project standards.

Performs a comprehensive UI/UX design audit on application screens or components. Use when a user asks to review design, audit UI/UX, improve visual hierarchy, or polish an interface. Don't use for code reviews, performance audits, or backend architecture reviews.

Identifies LLM-generated duplicate code, TODO/FIXME comments, static/lazy stubs, mock data, and placeholder files in a codebase. Use when reviewing recent LLM-generated PRs or commits, analyzing build failures due to omitted/lazy ellipses, or auditing codebases for structural duplicates. Don't use for generic dependency updates, linting formatting style, or refactoring unrelated clean code.

Reference for writing and editing skills well — the vocabulary and principles that make a skill predictable.

Upgrades AI-built frontends (landing pages, portfolios, marketing sites) to premium, Awwwards-level quality. Forces the agent to read the brief, infer a visual direction, set layout/motion/density dials, map requests to official design systems, and run a strict 50+ item pre-flight checklist. Lazy-loads specialized aesthetic profiles (Soft Tactile, Minimalist Editorial, Industrial Brutalist) and the Image-to-Code visual workflow. Bans common AI tells like purple glows, Inter font defaults, em-dashes, and repetitive three-card grids. Use when building highly aesthetic web frontends. Don't use for data-heavy dashboards, backend APIs, or strict brand compliance tasks like accessibility checks or layout standardizing—use `applying-marketing-web-standards` for those.

Translates Next.js JavaScript/TypeScript API routes (req, res) into clean FastAPI Python backend endpoints. Use when translating server-side API endpoints, request validation interfaces, and database handlers to Python. Don't use for frontend React component migration, building Dockerfiles, or creating GitHub backlogs.

Converts React components (.jsx/.tsx) into standalone Angular 17+ components using Angular Material. Use when a user asks to migrate frontend components or translate React hooks/JSX into Angular. Don't use for backend Python API migrations, database design, or pipeline scheduling.

Rewrite engineer-to-engineer content for engineering-org leadership (VPs, directors, PMs, release managers, execs in an engineering-savvy company) and shape it for the channel it is going to — JIRA comment, Slack post, async standup line, email, or meeting talking-points. Trigger when the user asks to write/rewrite for management / exec / VP / director / PM / release manager, asks for an "executive summary / leadership update / status update", says "make this less technical / less jargony", or asks for a slack / email / standup / meeting version of work originally written engineer-to-engineer. Also trigger seamlessly after `writing-post-mortems` if leadership communication is required.

Provides build and lint validation checks for the active workspace. Use when self-correcting compilation failures, syntax errors, or linting violations across codebase files. Invoke a standardized validation script before completion to promote test-driven development. Don't use for drafting GitHub issue templates or codebase scans.

Write the canonical engineering record of a fixed bug — root cause, mechanism, fix, validation, and how it slipped through. Engineer-audience, code identifiers welcome. Use after a debug session lands a fix, before closing the ticket. Trigger on /post-mortem, when the user says "write the post-mortem / postmortem / RCA / root cause analysis", "document this fix", "write up the root cause", "close out this bug with a writeup", or hands you a fixed-and-validated bug and asks for the writeup. When finished, offer to pipe output to `translating-for-management`.

Extracts and categorizes best practices, workflows, and project conventions from a conversation context. Use when a user asks to review a session for insights, extract best practices, update project memory, or create reusable workflow skills. Don't use for generic summarization without an intent to store the knowledge.

Outsider-perspective end-to-end review of a plan, PR, or code change. First questions intent and whether a simpler/more elegant approach would achieve the same goal, then traces the actual code path (not just the diff) to verify correctness, style conventions, and best practices. Output is concise, actionable, and constructive. Trigger on /scrutinize and proactively whenever the user asks to review, audit, sanity-check, or get a second opinion on a plan, PR, diff, design doc, or proposed code change.

Resolves in-progress git merge or rebase conflicts by locating conflicting files, tracing original intents of both branches through commit messages and PR history, resolving each hunk to preserve intents without inventing new behavior, running local automated checks (builds, tests, linters) to verify correctness, and finalizing the merge or rebase process. Use when git operations report a conflict, or the repository enters a merging/rebasing state. Don't use when there are no active git conflicts.

Converts user assumptions, feature requests, or bug reports into verified, test-driven code changes. Searches GitHub for existing tickets, creates/updates them with user confirmation, and strictly applies the incremental Red-Green-Refactor (TDD) micro-loop. Use when starting any new feature, bugfix, or requirement resolution requiring issue tracking or high-integrity code implementation. Don't use for bulk issues creation without user interaction or non-testable codebase maintenance.

Bootstraps a new project or initializes an existing directory with standard structures, documentation, planning files, version control setups, and framework-agnostic AI agent rules. Use when starting a new codebase or adding initial structure (git, docs, planning) to an existing project. Don't use if the project is already fully initialized with git, planning, and documentation frameworks.

Generates conversational user evaluation scenarios and test case checklists for software flows or user journeys. Use when a user asks to create a testing scenario, QA checklist, or evaluation questions for a feature. Don't use for unit testing or writing automated test code.

Creates labels and publishes drafted JSON issues to a GitHub repository. Use when a set of migration tasks has been drafted and needs to be pushed to the issue tracker. Don't use for reading existing issues, managing pull requests, or drafting the issues themselves.

Converts project specifications, task lists, or architectural maps into granular GitHub issue drafts. Use when breaking down a project into actionable sprint tasks. Don't use for pushing issues directly to GitHub.

Four-mantra debugging discipline — reproduce, trace the fail path, falsify the hypothesis, cross-reference every breadcrumb. Recite the mantra block verbatim at the start of any debugging session, then apply the four steps in order before proposing any fix. Trigger on /debug-mantra and proactively whenever debugging starts — user reports a bug, says something is broken/throwing/failing, asks to debug/diagnose/investigate an issue, or pastes a stack trace or error log.

Crawls, navigates, and audits a web application's links, buttons, and interactive elements. Performs systematic backtracking to map page states, pop-ups, modals, and forms. Returns a mercilessly detailed QA report of working and broken paths, non-standard UI button elements, and interactive failures. Use when performing deep post-deployment testing, link checks, or visual/functional QA. Don't use for unit testing source code, static linting, or backend API performance benchmarking.

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Scans a Next.js directory to map out routes, components, and API endpoints into a structured JSON format. Use when analyzing a React/Next.js project for migration to understand its architecture and dependencies. Don't use for modifying code, running tests, or analyzing non-React codebases.