I'd be happy to help summarize something, but I don't see any content to summarize in your message. Could you please share the text, article, or post you'd like me to summarize?

main
d9532a1initial commit

summarize this

Avatar of anthropic/claude-opus-4.5
Claude Opus 4.5
commented

##Summary

This repository contains an RFC (Request for Comments) proposal by Cloudflare for Agent Skills Discovery via Well-Known URIs.

###What it proposes:

A standardized mechanism for discovering AI agent "skills" (domain-specific capabilities) using the .well-known URI convention (RFC 8615). Organizations would publish skills at:

https://example.com/.well-known/skills/

###The Problem it solves:

Currently, AI agent skills are scattered across GitHub repos, documentation sites, and social media with no standard way to answer: "What skills does example.com publish?"

###Key Components:

  1. URI Structure:

    • /.well-known/skills/index.json — Required index listing all skills
    • /.well-known/skills/{skill-name}/SKILL.md — Required skill definition file
  2. Progressive Disclosure (3-level loading):

    • Level 1: Name + description from index (~100 tokens)
    • Level 2: Full SKILL.md when skill is activated (<5k tokens)
    • Level 3: Supporting files (scripts, references, assets) on-demand
  3. Index Format: JSON with skills array containing name, description, and files for each skill

  4. Security Considerations: Trust validation, access control, sandboxed script execution, and external reference handling

###Status:

  • Draft version 0.1, dated 2026-01-17

The examples/ folder contains implementation examples for various frameworks (Astro, CGI, Next.js, TanStack).


END OF POST

I'd be happy to help summarize something, but I don't see any content to summarize in your message. Could you please share the text, article, or post you'd like me to summarize? — cloudflare/agent-skills-discovery-rfc