Long-term memory for Codex: preserve decisions and context across chats in a local-first Obsidian/Markdown knowledge base with staged writes you approve and index + graph retrieval. Karpathy-style, OKF-inspired.

Anti-bloat prompt linter for Claude Code, Cursor & ChatGPT. Flags prompt-engineering bugs against a stable rule catalog (PR001-PR-INJ03). Lint, don't rewrite. No framework imposition, no interview flow, no language switching.

Open-source Cursor skills for cold email outreach — deliverability, copywriting, and Instantly CLI

Inspect, visualize, and generate action-flow graphs for Codex-style agent skills with a local web UI and bundled graph generator.

Cross-agent skill for private Obsidian sync across Windows, GitHub, iPhone, and Android

Zero-dependency in-browser WYSIWYG editor you can bolt onto any HTML slide deck — inline text edit, add images, drag/resize, undo/redo, slide reorder, autosave & export. Pairs with html-ppt/reveal decks.

A Lovable-inspired design skill for AI coding agents to create distinctive, production-grade frontend UI.

A Claude Code skill set that makes Claude argue with itself before it agrees with you

Codex Skill for converting Axure/HTML prototypes into editable Ardot files / 将 Axure/HTML 原型转换为可编辑 Ardot 文件的 Codex Skill

Agent Skills for Claude Code — scaffold Home Assistant add-ons, review MCP servers, and more. Installable as a plugin marketplace.

A curated catalog of self-improved agent skills for Hermes and compatible agents

AI-powered Xiaohongshu content topic planner skill

AI Agent Skills for planning and writing cold-email campaigns.

Minecraft AI agent skills bundle for Java and Bedrock server administration, development, packs, crossplay, and automation across Codex, Claude Code, and compatible agent hosts.

Verify AI agent output with actionable evidence for non-engineers. Build better prompts by auditing AI claims and checking task status before continuing.

Fable5-like evidence-first Codex skill for agent workflows, verification-first delivery, adversarial review, and portable prompts.

Render agent context as images. Same text, ~60–75% fewer tokens.

CLI, SDK, and gateway for Qdrant — SQL-like queries, score shaping, retrieval diagnostics, and stable JSON output for scripts and agents.

Agent-facing MCP and skills for DSView/sigrok logic analyzer workflows

Portable Agent Skill that saves a lean session note at the end of a coding-agent session, with a one-shot installer for Claude, Cursor, Codex, Pi, and opencode and your choice of knowledge base (Obsidian, Notion, or any folder). Comes with a Marauder's Map.

Build approved, bounded Codex sub-agent loops with role isolation, deterministic validation, and evidence-locked execution.

Description: A structured engineering workflow skill for complex AI-assisted coding, coursework, data analysis, and research projects.

Plan everything before anything. Single-command AI agent skill that emits a complete, audit-proof, agent-executable master plan (PLAN.mdx) before any code is written. Claude Code, Codex, Cursor, OpenCode, and every Agent Skills client.

Carinya Parc Agent Skills

Visual-director Agent Skill for 2K Chinese Remotion videos with original imagery, animated text and automated QA.

Useful Skills from Bogdan Baciu

基于工作区文件传递的模块化 Agent Skill 金融分析工作流

Author once, validate once, build skill packs for multiple coding agents.

Agent skill engine for AI-tool affiliate catalogs, monetization audits, partnership workflows, and revenue loops.

Score your own Claude Code sessions on YC Paxel's 5-axis builder rubric — locally, with Claude Haiku as the scorer. No upload, no backend.

Claude Code skill for PhD-level econometrics (OLS, IV, DiD, RD, panel FE, MLE, STATA) distilled from a Korean lecture series.

Creating, writing, or scaffolding a SquadJS plugin, adding a server-side feature to a Squad server via SquadJS, or deciding whether SquadJS can support a desired behaviour

Agent skill that picks the right AI model using live benchmarks from artificialanalysis.ai. Works with Claude Code, Cursor, Codex, and any skills.sh-compatible agent.

9-verb knowledge engine for Obsidian vaults — Karpathy LLM Wiki pattern. Works with Claude Code, Antigravity, Codex, Gemini CLI, Cursor via symlink distribution.

Reusable AI agent skills, project templates, and guardrails for safer software maintenance and delivery.

Open source agent skills for evidence-first startup ideation: capture pains, grill assumptions, design validation tests, synthesize interviews, and scope a narrow v1.

Enables AI agents and developers to run WP-CLI, mysql, and composer in LocalWP site environments on macOS with zero configuration.

Agent skills for Unreal Engine + AngelScript workflows. Drop-in skills.sh-compatible monorepo.

Claude Code/Codex plugins for LlamaParse, LiteParse and LlamaParse MCP

CLI for iCloud Hide My Email — for humans and AI agents

Price on the Proof — a Claude skill for pricing consulting, coaching, workshops, and courses. By Chris Gee.

Rust-powered video analysis skill for Hermes Agent — transcript-first with scene detection

Agent skills, diagnostics, and Designer MCP guidance for safe SeatLayer integrations across Web, React, Flutter, and iOS.

Claude Code skill for building production-grade, visually distinctive frontends — design brief, component patterns, animation library, and self-evaluation rubric in one unified workflow.

Agent skill for reviewing code or docs

SST3 Skills — Claude Code plugin marketplace. Ralph Review Trio (Haiku → Sonnet → Opus, restart-on-fail) + more packs coming.

Audit agent skills for CLI drift: find skills that call external tools with flags/subcommands the installed version no longer accepts.

Personal collection of Anthropic Agent Skills with capability + trigger evals — currently: ngrx-signals (Angular Signal Store, +16pp pass rate vs baseline)

Crack a single stock for an investment view — thesis-first equity research workflow for Claude Code. 13 phases from raw filings to a defensible 5-8 page investment memo. Direction commits before the model is built.

Intelligence pack for software engineering based on Ainova Systems best practices

Standalone governed Proxmox VE ops — 23 MCP tools with built-in audit/budget/undo/risk-tier harness

Executable Agent Skills for PR, SEO/GEO, paid media, Xiaohongshu full-funnel operations, content, brand and global growth｜WEPR 可执行增长技能库

Polywave implementation for OpenAI Codex CLI: parallel AI agents with disjoint file ownership and worktree isolation. Requires polywave-tools.

Make AI-generated Three.js / 3D actually look right — a portable skill that fixes the wrong scale, backwards models, floating objects & broken cameras your AI ships. Works with any AI assistant.

A Claude Code / Codex skill that turns any complex topic into a 24-hour active-learning sprint.

Universal E2E & manual acceptance testing skills for Claude Code — every run writes a protocol

tui use for agent