🎯
Skills
Loading...
Showing 30 of 145012 results
A shared agent skill that implements approved PRODUCT.md and TECH.md specification files while keeping specs and code aligned, part of Warp's common-skills collection.
A shared agent skill that provides guidance for creating and maintaining skill directories and SKILL.md files, part of Warp's common-skills collection for skill authoring.
A shared agent skill that provides a workflow for inspecting GitHub CI failures and producing a fix plan, part of Warp's common-skills collection of reusable development workflows.
Safe, documented, reversible releases.
Production-grade engineering skills for AI coding agents that encode senior-engineer workflows, quality gates, and best practices across the full software development lifecycle from spec to shipping.
Quality gates before merge.
A shared repository of reusable agent skills for cross-repo workflows, covering spec writing, PR creation, CI failure diagnosis, code review, merge conflict resolution, and skill authoring.
Transforms AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Features WAL Protocol for corrections, Working Buffer for context survival, and Compaction Recovery.
Semantic design system skill that generates DESIGN.md files for Google Stitch screen generation, encoding visual atmosphere, a banned-color list (no AI purple/neon), typographic architecture (no generic Inter/Times), asymmetric layout rules, and perpetual micro-motion — so Stitch outputs premium, non-generic UI instead of the default AI-slop look.
A production-grade agent skill from addyosmani/agent-skills that activates the /code-simplify command to guide AI agents toward clarity over cleverness when refactoring code.
Contact (通讯录) skill of the WeCom (企业微信) CLI — lets humans and AI agents fetch the visible-scope member list and search members by name or alias from the terminal via `wecom-cli contact ...` commands.
Small slices, right context, clean interfaces.
A Flutter animation skill from the MAD Agents Skills collection, providing structured knowledge on animation APIs, patterns, and best practices based on official Flutter documentation for AI-assisted development.
Automatically diagnoses and repairs broken OpenCLI adapters when a site's DOM, API, or response schema changes — only touches the adapter at `RepairContext.adapter.sourcePath`, enforces hard stops for AUTH_REQUIRED / BROWSER_CONNECT / CAPTCHA, a 3-round repair budget, and files an upstream GitHub issue after a verified fix.
Part of the dbskill business diagnostics toolkit distilled from 12,307 tweets. This skill helps migrate any project into a Claude Code/Codex-compatible agent workspace by auditing rule files, identifying source of truth, and unifying naming conventions.
Prove it works with tests and real browser data.
A production-ready plugin system with 112 AI agents, 146 skills, 16 workflow orchestrators, and 79 development tools organized into 73 focused plugins for Claude Code.
Generates AI chat interface components following shadcn/ui and Vercel AI SDK conventions for faster AI application development.
Xiaohongshu (小红书) title formula skill that picks the right formula from 75 verified viral templates across 12 psychological-trigger categories (cognitive conflict, curiosity gap, fear/loss, identity, number anchoring, outcome promise, social proof, controversy, context, call-to-action, authority, interaction). Generates 5–8 tailored titles per topic, each traceable back to a specific formula ID with the original viral reference and rationale.
A Tavily skill that performs comprehensive AI-powered web research with citations, enabling agents to conduct deep multi-source investigations and produce well-referenced research reports.
Diagnoses and resolves Vue 3 runtime errors, async failures, hydration issues, and common component debugging challenges.
Prove it works with tests and real browser data.
Guides the creation of a PRD (Product Requirements Document) through a structured interview process covering problem definition, solution options, scope negotiation, deep module extraction, and user story generation, then submits the result as a GitHub issue.