A persistent, unified memory layer for all your AI agents (e.g. Claude Code, Codex), backed by Markdown and Milvus.

The definitive resource for Agent Skills - modular capabilities revolutionizing AI agent architecture

Claude Code skill for optimizing oversized CLAUDE.md files using progressive disclosure

Palestra técnica sobre Spec-Driven Development (SDD) com GitHub Copilot

Cross-agent skill quality gate for SKILL.md files. Validates frontmatter, scores description discoverability, checks file references, enforces three-tier token budgets, and flags compatibility issues across Claude Code, VS Code/Copilot, Codex, and Cursor.

🌿 Prune your AI agent's context window. Reduce token usage by 70-95% with hierarchical memory. Replace flat MEMORY.md with a bonsai-shaped domain tree. Progressive disclosure, zero dependencies. Works with OpenClaw and any LLM agent framework.

🚀 Explore modular agent skills for dynamic AI development, enabling on-demand knowledge injection through standardized `SKILL.md` packages.

Progressive file disclosure for agentic AI - explore files hierarchically with plugin-based analyzers

Progressive disclosure retrieval for long-horizon AI agents over personal knowledge bases. Three-tier model: hot cache → FTS5 index → full read. Local-first, zero dependencies.

Desktop app (Tauri 2 + SolidJS) to compose Claude Code workspaces with curated skill packages, MCP servers, hooks, commands & CORE-documenten — OpenAEC Foundation

🛠️ Discover essential skills and tools for AI coding agents to enhance capabilities and streamline the development of intelligent solutions.

Persistent project-scoped knowledge base for Claude Code and other SKILL.md-compatible code CLIs: deep research findings with progressive disclosure and contrarian-pass investigation. Survives /compact. Installs via npx skills add, /plugin install, or git clone.

Claude Code Plugin for Multi-session AI development with smart context loading

An AI skill for estimating AI-assisted and hybrid human+agent development work. Research-backed formulas. PERT statistics. Calibration feedback loops. Zero dependencies.

Generate progressive-disclosure interfaces from complex APIs via My-Lang

Section-level SQLite library for Claude Code skills — search, checkout, and compose any part of any file with 99% token savings.

Ecosystem-aware CLI for validating and upgrading Agent Skills: enforce spec compliance, detect vendor extensions without breaking, convert large skills into progressive disclosure.

A Markdown-based task runner — documentation first, commands second.

A Claude Code skill that delivers information progressively instead of all at once

Agent Skills 学习工坊 · 基于 agentskills.io 官方规范整理的 55 个核心知识点（SKILL.md、YAML frontmatter、渐进式披露、最佳实践、Evals、客户端接入、生态与安全）