21st
claude21st.dev UI skills and MCP. Pinned upstream: 21st-dev/claude-code-plugin.
7 skills · 1 MCP server
How work is run: best-practices, learning-loop, model-delegation, visual-loop, token-efficiency, actions-cost, netlify-cost, session-continuity, delivery-loop, knowledge-base-builder, plugin-experience-design, product-passport, and the cheap-reviewer subagent.
Plugin claude-workflow from marketplace ogfrench in ogfrench/agentic-coding-setup. Commands read from the repo's manifest — today.
actions-costReduce GitHub Actions costs while preserving PR checks, security and deployments.
architecture-gardenerInspect architecture drift, ownership boundaries and recurring workarounds. Use for architecture audits and structural root-cause repairs, not routine edits.
best-practicesLearned practices for tests, data, security, UI, CI and agents. Use for domain changes or best-practice questions.
delivery-loopFinish authorized repository tasks through code, tests, CI and PR review. Use for end-to-end delivery and release handovers.
design-system-adapterApply an existing visual identity or design system to a slide deck, page, chart or prototype. Use for faithful brand adaptation and visual consistency, not as a required preliminary stage for design.
diagram-craftCreate editable, source-grounded Mermaid and clean diagrams. Use for architecture, components, processes and sequences.
issue-managementTriage, write and reorganize GitHub issues and roadmaps with MECE scope, one owner per outcome, clear decisions and verifiable acceptance checks. Use for backlog cleanup, feature planning, issue splits and duplicate audits.
knowledge-base-builderBuild or audit agent-ready knowledge bases over code and documents. Use for knowledge bases, project brains, hybrid retrieval, RAG, MCP search tools and source-grounded answers.
learning-loopCapture and promote reusable lessons into skills and best practices. Use after bugs, tool quirks, corrections and milestones, or when asked to remember a lesson, make a skill, promote a learning or harvest learnings.
model-delegationChoose the main agent, a reusable skill or tool, or a subagent using measured quality, context and cost. Use for delegation, model splits, agent workflows, orchestration design, cheaper models, and latency optimization.
netlify-costEstimate Netlify cost before changing its config or build.
plugin-experience-designDesign and validate AI plugin downloads, installation and onboarding for Claude Code/Cowork, ChatGPT and Copilot. Use for host support, editions and release readiness.
presentation-designCreate or revise finished presentation slides, decks, charts and visual explanations in supported formats. Use for slide production; storytelling assistance is optional and proportionate.
presentation-storyDevelop or critique presentation narratives, arguments, slide sequences and action titles when story help is useful or requested. Not a prerequisite for slide creation.
product-passportMap product features to code and verify real user journeys. Use for product maps, cross-layer regressions and release evidence.
session-continuityPreserve PR checklists and evidence through compactions and multi-agent handoffs. Use for session continuity, progress checkpoints and handover.
token-efficiencyFewer turns and tokens for the same work: batching, lean context, background waits, facts saved once.
visual-loopInspect real rendered UI, compare design options and catch visual defects. Use for screenshots, redesigns and visual reviews.
cheap-reviewerRead-only correctness and safety review of one commit, or of the working diff, before it is pushed. Give it the commit (or "the working tree"), what the change is meant to do, and the risky spots to look at. Returns a short ranked list with file:line and a concrete scenario for each finding. It neve
21st.dev UI skills and MCP. Pinned upstream: 21st-dev/claude-code-plugin.
7 skills · 1 MCP server
Portable companion for setup auditing, adversarial review, real UI evidence and release handover.
4 skills
Writing skills: frogify (frog voice), deslopify (copy audit), humanizer (remove AI tells).
3 skills
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.