Agentic Skills
Open-source collection of Agent Skills in the agentskills.io open format for AI agents like Kiro, Cursor, Windsurf, Claude Code and others. Each skill is a reusable module that teaches the agent to perform complex tasks with context, structure and best practices. Born from the real need to direct agents consistently.
Agent Skills are an open and lightweight format for extending AI agent capabilities with specialized knowledge and workflows. With the rise of coding assistants and autonomous agents, I realized that directing these agents consistently required well-structured instructions. Each skill emerged from a practical need in real projects: standardizing outputs, optimizing for Generative Engines, generating premium proposals, or creating coherent designs.
A public repository where any developer can find ready-to-use skills for AI agents. Currently features skills like Premium Proposal Builder (creating proposals, slide decks and sites optimized for purchase decisions) and GEO Optimization (optimizing content for Generative Engines to maximize citations in AI responses). Each skill follows the open agentskills.io specification with configuration frontmatter and progressive disclosure.
Creation, curation, and maintenance of all repository skills. Architecture definition following the Agent Skills specification, bilingual documentation (PT-BR and EN), practical testing across different agents (Kiro, Cursor, Windsurf, Claude Code, OpenCode, Codex), and continuous evolution based on real project needs.
Skills structured in Markdown following the open Agent Skills specification, with a SKILL.md file containing metadata and instructions that agents load on demand via progressive disclosure. Installation via CLI with npx skills add fabricioctelles/skills or agent-skills-cli. Compatible with Kiro, Cursor, Windsurf, Claude Code, OpenCode, and Codex. Apache 2.0 license.
Public and active repository on GitHub with two published skills (Premium Proposal Builder and GEO Optimization) and constantly evolving. Next steps: expand the collection with new specialized skills, refine existing ones based on community feedback, and follow the evolution of the Agent Skills ecosystem.