design
claudecodexDesign skills — UX psychology, visual design, accessibility, and Apple featuring readiness for web, SaaS, and mobile apps
4 skills · 2 agents
Every plugin for OpenAI Codex on GitHub — repos with a .codex-plugin/plugin.json or a Codex marketplace. Install with codex plugin marketplace add owner/repo then codex plugin add name@marketplace. Many work in both tools; see Claude Code plugins.
Design skills — UX psychology, visual design, accessibility, and Apple featuring readiness for web, SaaS, and mobile apps
4 skills · 2 agents
GitLab MCP server — issues, merge requests, pipelines, code search. 80+ tools.
6 skills
Forces genuine simplicity over patchwork. Fix the pipe, not the floor.
5 skills
Build, refine, and analyze Tough Tongue AI voice agents for tough conversations — agents that automate them (sales calls, lead qualification, demos, screening, booking) and agents that train you for the ones you handle yourself (negotiations, interviews, coaching). Create scenarios, place phone calls, schedule meeting bots, and analyze sessions.
4 skills · 1 MCP server
Evaluate AI skills across model tiers with blind testing and refinement recommendations
1 skill · 1 command · 2 agents · hooks
Teach your coding agent to write Cities: Skylines II code mods: a working loop that checks the result in the running game, and the traps an agent with the decompiled game open still falls into. Ships the cs2-dev MCP server, which restarts, stops and awaits the game, reports its errors, and runs a mod's in-game scenarios.
4 skills · 1 MCP server
Tandoor Recipes MCP server: full Tandoor API coverage with resources, prompts, and AI URL/image import.
4 skills · 1 MCP server
docent — narrated, animated explainer films for any subject (a codebase, a PR, an essay, a wiki, a URL). Four mode-centric slash commands: /docent-doctor (first-run bootstrap and health check), /docent-pr (PR-review film), /docent-ar (architecture-review film), /docent-explain (the one-shot — any subject, any mode, end to end). Use when the user asks to review a codebase as a film, explain a subject as a video, review a pull request at depth, or hands over a subject expecting an artifact at the end.
4 skills · 1 agent
Turns lecture slides, labs, meeting notes and papers into a linked Obsidian study vault: atomic notes with hover definitions, embedded slides, comparison tables and cross-course bridges.
1 skill · 4 agents
Deterministic vector motion engine, virtual-clock studio player, and 1080p 60fps MP4/WebM launch-video compiler for AI agents.
4 skills · 1 command
Build a local source-backed recall context runtime for Claude Code agents.
3 skills · 1 MCP server
Code quality skills — refactoring, naming conventions, project structure, and internationalisation
4 skills
Design, implement, and review web interfaces with shared UI guidance
2 skills · 1 command · 1 agent
Live-edit open PowerPoint presentations via Office.js MCP bridge
2 skills · 1 MCP server · hooks
Visual markup on a running dev server, handed to the agent as one locked batch. Registers Tailr's MCP server and the operating rules for the review loop.
3 skills · 1 MCP server
Advisory evidence reminders for technical and data work, targeted ambiguity checks, claim fidelity, and self-audit.
3 skills · hooks
Skills for interacting with 1Password through the op CLI
4 skills
Say what to watch; a small probe checks it in the background, and Claude comes back to your session when it fires.
2 skills · 1 MCP server · hooks
Evidence-packet linter for claim-to-source references, quotes, lexical support, and unmatched numbers, plus compatibility agenda-analysis skills.
3 skills · 1 MCP server
在 Codex 里用 LibTV 做视觉创作。
3 skills · 1 MCP server
Convert files between 140+ formats, pull data out of documents, and build custom converters, right inside Claude.
3 skills · 1 MCP server
Describe the effect you want and EffectHouse AI builds it in the real Effect House engine: face and hand tracking, stickers, makeup, 3D, text and scripted mini-games. Watch it in a live preview or on your phone, tune it with sliders, then publish it to your Effect House account and follow how it performs.
1 skill · 2 MCP servers · hooks
Connect Claude Code to a local Froglet node for verifiable compute, signed receipts, settlement, and service publication.
1 skill · 1 command · 1 MCP server
Measured image optimization (PNG, JPEG, GIF, SVG, WebP, AVIF) with quality profiles, banding checks and a 1:1 review.
2 skills · 1 command
Disciplined brainstorming, implementation planning, and plan-aware code review skills.
3 skills
Choose the best TrustedRouter model using live price, speed, quality, privacy, context, and provider-health data.
1 skill · 2 MCP servers
Aomi for AI agents — drive the Aomi CLI from natural-language prompts (chat, simulate, sign on-chain transactions with account-abstraction-aware execution) and scaffold new Aomi apps from API specs. Bundle ships two skills: aomi-transact (run on-chain across 40+ DeFi/perps/CEX/social apps on EVM mainnets and L2s) and aomi-build (scaffold Rust SDK crates from OpenAPI/Swagger specs).
1 skill · 2 MCP servers
Prepare, review and reconcile user-authorized Aomi transactions through durable Commits, with operational funding checks and explicit receipt/recipient verification.
1 skill · 2 MCP servers
Provides skills to initialize or reconcile project documentation (README.md, CLAUDE.md, subsystem docs)
3 skills
felt — fiber-based thinking substrate. Bundles the felt skill (substrate practice), the shuttle skill (autonomous dispatch practice), and SessionStart + PreToolUse hooks that deliver active fibers and gate non-felt tool use until the felt skill is activated.
2 skills · hooks