Practical agent skills for projects, decisions, business, research, and communication

开源论文引用核验 Skill：检查 DOI、PMID、BibTeX、参考文献真实性和论断支持关系。

The map: how the runsagents artifacts compose into one control loop for agents with production stakes. A control loop, not a toolbox shelf.

Records agent handoffs as provenance records with task state, ownership, and verification evidence.

通用 Agent Skill：执行前发现未知、执行中记录偏差、完成后确认理解｜源自 AI Builders 解读 01

Claude Code skill — direct Trello API access via curl. Read cards, post comments, upload files, download attachment bytes. No MCP server needed.

Claude Code skill that reports local token usage (no API key, no network) for Codex, Claude Code, Pi, Grok, opencode, and openclaw sessions

A Claude skill that builds truthful, ATS-friendly, impact-oriented resumes — no invented metrics, no inflated ownership. Interviews you for evidence and marks what needs verifying.

PowerShell skills and guardrails for reliable Windows command execution by AI agents, with Codex and Claude Code adapters.

Fetch web content into a named local cache for multiple reuse. Convert HTML to Markdown by default, refresh expired entries, and maintain a JSON manifest.

Local, on-device audio toolbox for AI agents: voice-clone TTS, instrumental music, and audio verification — three installable Agent Skills.

Operating discipline for rigorous, reproducible computational-science research — a portable, markdown-only Claude skill pack: 18 disciplines, a generator, and read-only auditor agents.

Public agent skills for OpenCode, Claude Code, Cursor, and other agent runtimes

A Claude skill that reviews an AI agent for production readiness across evaluation, cost, observability, guardrails, human oversight, reliability, and governance.

Agent skill for personal finance — log expenses, import statements, track budgets and trades via the Orchune MCP server. Works with Claude Code, Hermes, and OpenClaw.

嘴替·催婚包 —— 过年回家最难回的「你到底怎么想的」，得体/有针/有底气三种回法；先分清是套路施压还是焦虑的爱。Made with 嘴替.skill

Portable Agent Skills for Fable 5-aware workflow routing

40 portable full-SDLC agent skills for Claude Code, Cursor, Codex, and GitHub Copilot — deep-work autonomy, contract-guard for external interfaces, TDD, code review, PR babysitting. Install: npx github:IcodeNet/agent-skills

Claude skill: audit your project's public pages the way a first-time visitor experiences them — every page read fully, every link clicked and counted.

Multi-agent PR review skills for Claude Code: reviewer-ultra (regressions/bugs/smells + root-cause adjudication) and readable-ultra (readability review from Clean Code principles)

Curated, auditable, benchmark-ready Agent Skills library for Claude Code, Codex, OpenCode, Cursor, and more.

Cross-Harness operations for Agent Skills: task preflight, missing-Skill discovery, local evidence, and reversible governance.

Evidence-backed agent skill for distinctive web design, frontend development, audits, and production delivery.

LLM-first Go CLI for managing multiple Claude Code OAuth sessions

AI coding agent skill for turning hidden unknowns into actionable knowns before implementation.

Compile code for machine attention with a lazy, content-addressed semantic IR.

Say what you need — Skill Scout finds, vets, installs & runs the right skill, with the code reviewed before anything installs. For Claude Code, Codex & Cursor.

Agent skill for the Pi coding agent: async cmux sub-agent delegation over intercom. Maintained by ember-skua-45.

Domain-agnostic Claude Agent Skill: strategic challenge phase + self-verifying execution loop (Arabic + English) | مهارة الوكيل الاستراتيجي ذو الحلقة المغلقة

One canonical Agent Skills tree, symlinked into every agent CLI — edit a skill once, Claude Code, Codex and friends all see it

Git-backed version control, rollback, and provenance for your Claude Code & Codex Agent Skills — a safety-gated skill package manager (CLI + plugin).

An adaptive interview Skill that clarifies ambiguous ideas through one high-value question at a time.

人类理解优先的 Obsidian 知识库 Skill：从外部资料到阅读卡、个人批注、可追溯 Markdown 与可持续维护的 wiki。

Agent Skill for creating privacy-safe public pull requests with redacted history and worktree audits.

A sanitized snapshot of my Codex agent stack: skills, plugins, MCP services, tooling, and operating rules.

ぬか喜び検出器っぽいものが本体です。Home of the False Hope Detector Skill — pre-investment kill tests for weak ideas.

Cursor/Claude agent skill: numbers-first IP due diligence — prior art, Gold Index, FILE/BORDERLINE/DO NOT FILE PDF dossier.

Public Codex plugin marketplace for Business Central Assisted Setup development.

Clone reference designs, reverse-engineer the design thinking, customize conversationally — a design plugin for Claude Code, Codex, Cursor, and OpenCode.

Cross-agent 信达不雅 translation Agent Skill; target language agnostic, defaults to Chinese

A fast, offline static risk analysis CLI for AI agent skill files. Detects malicious instructions, steganographic payloads, and dangerous capability chains.

Micro-pacing cognitive prosthetic. Stops an AI agent from stacking multiple decision questions in one turn.

C.A.S.H. (Constraints, Ambiguity, Sequencing, Heterogeneity) is an agent skill and Orchestration Router that evaluates system components and routes them to traditional code, an LLM call, or a hybrid.

Claude Code skill to title, name, and label your conversations - interactive, multi-variant, multilingual titles. 🏷️

Collection of skills for agents

A director-level Codex Skill for original multi-shot AIGC short films.

Reusable Codex and Claude Code skills for research operations, grants, evidence synthesis, lab-in-the-loop planning, file curation, and faculty workflows.

Create reusable AI audit, quality-review, compliance-review, content-moderation, customer-service QA, or order-review automation plans

Agent skill orchestrating a full frontend+design workflow: research → output → anti-slop review → stack/vocab reference. A router, not a monolith.

Meta-skill for Claude Code: discovers installed skills, decomposes multi-domain tasks, sequences skills in dependency order, and synthesizes one cohesive deliverable

Agent skill to install, update, and verify project skills from pinned, content-hashed sources

Built for vibe coders and lazy prompters: inverts short, unpolished prompts into Principal-level engineering directives aligned to the right discipline, then executes them. Claude Code plugin + Codex CLI skill.

Commander-intent gate for autonomous, verified human-agent execution loops.

Agent Skill to convert CSV and LaTeX result tables into editable publication figures.

Personal Codex and agent skills collection

Open-source Claude Code plugin and Agent Skills for TanStack Start: type-safe routing, server functions, SSR, middleware, authentication, testing, and deployment.

Agent skill: spawn and manage a disposable subagent team for any mission — capability probe, right-sized topology, mission briefs as contracts, research-first, plan gate, and an integration gate that verifies every member's work before merge

Agent skills, Rust CLIs, and Python jobs for a content-marketing ops fleet — maker/checker gates, curation metric, workload ledger, ops-exhaust content machine, skill-tuner.

Hand coding work to Codex without handing over control. Fable architects and reviews; Codex builds in an isolated worktree.