Agent Skills made by Panno

Agent plugin for source-preserving academic de-AI and human-anchor-bound rewriting

One git checkout drops a shared AI coding setup (skills, subagents, MCP servers, OpenSpec scaffolding) into any project, across Claude Code, Kilo, OpenCode, Codex, and Copilot.

AI agent skills for App Store Optimization (ASO) and app marketing. 30 battle-tested skills for keyword research, metadata optimization, competitor analysis, and growth — powered by Appalize.

A registry of agent-loadable skills that distill technical knowledge and expertise from prior large-scale, enterprise engagements. Focus given to decision-making frameworks, trade-offs analysis, and addresses conditions that require subtle technical judgments. The Agent Skills open standard is used to develop all skills.

Agent-engineering patterns and portable, prompt-only skills for LLM coding agents — multi-agent orchestration, adversarial multi-LLM council, learned guardrails. Vendor-neutral, MIT.

Field survey of skill-repo layout patterns + BDD fixtures for lythoskill. Each stub demonstrates a real-world FQ-locator path style with self-check inside.

Agent-readable skills for bootstrapping and maintaining made.json feeds.

A collection of skills, plugins, and agents for AI workflows.

Production-Ready Agent Skills : product analytics, growth experiments, CRM, research synthesis, postmortems, data contracts, SaaS spend, compliance, architecture maps, and LLM eval and many more.

Code for SKILLS.re

A portable Claude Code plugin system for orchestrating software development through parallel specialist agents, enforced engineering standards, and automated quality gates. One command to plan. Parallel agents to build. Automated gates to ship.

Public agent skills for Scribeless handwritten mail workflows

Strict 3-stage handoff workflow (plan → execute → verify) for coding agents. Disk-backed state lets verify run in a fresh chat.

Personal collection of agent skills for use with Claude Code and other LLM agents.

Multi-agent NotebookLM research protocol with notebook-selection gating, dialectic verification gate, dual-output rule, and peer-lane collision prevention.

Forge portable coding-agent skills with linting, activation tests, render adapters, and local-first packaging.

A magical place where my skills, rules, and plugins for AI agents are defined, which magically boost productivity 🏰🪄

Agent Skill Infrastructure: quality check, behavior test runner, version awareness

Quiet, editorial chart PNGs for technical blog posts. 15 primitives, six themes, one CLI.

Collection de compétences agentiques modulables pour Kimi, Aegis et assistants IA

A curated pack of 7 practical agent-skills demonstrating real-world patterns: HTTP, GitHub CLI, ripgrep, file ops, JSON processing, encoding. Each SKILL.md validated against the v0.1 spec.

Critical-evaluation and R&D operating system: the governance and rubric layer other repos plug into.

Agentic Repository: Skills, Sub-Agents, Hooks, Context, etc.

Opinionated Claude Code scaffolder — agents, skills, hooks, slash commands. Like create-t3-app, but for your .claude/. Install months of Claude Code discipline in 30 seconds.

Frontend AI agent skills (SKILL.md) for UI pre-delivery checks, accessibility audits, and fixing patterns across Next.js, React, and other web projects

A curated collection of agent skills, designed and developed using Crush and Qwen3.6 27B.

A collection of AI skills created for me

This repository is a collection of custom GitHub Copilot skills, each in its own folder with clear specs, evaluation scenarios, and supporting scripts.

Three Codex skills for reflection, experience capture, and harness optimization

Agent Skills for Claude Code / agentskills.io-compatible agents. Cloudflare Workers + Discord integration patterns.

Reusable agent skills for Claude Code, OpenCode, Cursor, and the Vercel skills CLI.

Floop agent skills

Collection of skills

Public-record $1,000 → $10,000 automated trading experiment. Built by a founder and an AI as co-builders. Live SOL + ETH paper-mode on Coinbase perpetual futures.

Curated, sanitized export of 21 agent-skill packages for Claude Code and Codex, gated by an automated publication audit (no secrets, no local paths).

Create adaptive Codex-native agents with memory, onboarding, hooks, and dreaming.

A three-panel workspace that turns Claude Code into a tool for hard problems

Agentic skills; for hako code and other agents.

A collection of practical, production-ready AI agent skills for the skills.sh ecosystem.

Xcratch extension development skills for AI agents

Agent skills for the tools and workflows that I use.

This repository is meant to serve as a publicly available agent-skill to be toggled between to iterate on experimental agent-skills

tnagatomi's Agent Skills collection

Shareable skills and tools for AI agents

A collection of agent skills for repository setup, multi-CLI agent workflows, and intelligent context loading.

🚧 BETA · Skills for the handoff → compact → resume workflow. Preserve key session context across compactions, with a confirmation gate on resume.

Policy-as-code admission controller for AI agent skills and MCP tools. SkillBOM, lockfiles, and supply-chain baselines.

AI Knowledge Work Platform — Chat when you need answers, Work when you need results. Two modes, one seamless experience. Runs in terminal, browser, or desktop.

Agent Skill for MR/PR template previews and safe gh/glab submission workflows.

Agent skills for DOE work: branding, Empower (EVMS) reference docs, and PARS CPP JSON tools

Skills for writing and auditing CLAUDE.md files in derived/projected directories (symlink views, build caches, mirrors, generated docs). Skills.sh-compatible monorepo + Claude Code plugin.

Engineering discipline for AI coding agents. Spec-driven development, CI/CD, and documentation as composable skills built on the Workflow Router pattern.

Agent + étiquette: a small house of well-mannered agent skills for the skills.sh ecosystem.

Research-backed AI self-audit skill for Claude Code / Codex / OpenClaw. Verifies high-stakes answers via independent re-solve + cross-method probe — not by critique. Honest about limits; every design choice has a paper citation.