human-ai
Edit English prose to remove repetitive AI phrasing, inflated language, and mechanical structure while preserving meaning and the…
Sub-skill: Implement Personal Agent Protocol (Poppy) discovery. Publish /.well-known/poppy.json (draft 0.1) and validate its OAuth issuer binding. Informational check; does not affect level or score.
$ npx -y skills add fabricioctelles/skills --skill personal-agent-protocol --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/personal-agent-protocolContext preview
The summary Claude sees to decide when to auto-load this skill.
Sub-skill: Implement Personal Agent Protocol (Poppy) discovery. Publish /.well-known/poppy.json (draft 0.1) and validate its OAuth issuer binding. Informational check; does not affect level or score.
name: agent-ready-personal-agent-protocol description: > Sub-skill: Implement Personal Agent Protocol (Poppy) discovery. Publish /.well-known/poppy.json (draft 0.1) and validate its OAuth issuer binding. Informational check; does not affect level or score.
Publish public discovery metadata for the [Personal Agent Protocol draft 0.1](https://personalagentprotocol.org/docs/spec#3-discovery). This check is informational and does not affect readiness levels or percentage scores.
Serve JSON at `https://YOUR-DOMAIN/.well-known/poppy.json`:
{
"protocol_version": "0.1",
"organization": { "name": "Example Company", "domain": "example.com" },
"web": {}
}Replace the example organization with your actual name and domain. The domain must match the hostname of the requested well-known URL, ignoring leading `www.`. HTTPS redirects to a hosting provider are allowed, but that provider's hostname must not replace your organization domain. Return JSON, not your SPA's HTML fallback. Unknown fields and optional extensions are allowed.
Declare the interfaces you actually implement: `agent.protocols` entries with `type` and HTTPS `endpoint`, `apis` entries with `type`, HTTPS `url`, and `description`, or `web.browser_session_endpoint`. Agent/API declarations and browser-session endpoints require `auth.issuer`.
The issuer must publish RFC 8414 metadata with:
For issuer `https://auth.example.com/tenant`, publish metadata at `https://auth.example.com/.well-known/oauth-authorization-server/tenant`. Direct, device, and mediated sign-in declarations contain scope arrays. Mediated sign-in also describes its endpoint and credential fields; never put actual credentials or secrets in discovery metadata.
Publish only capabilities backed by working implementations. Passing this check confirms public discovery and issuer binding, not functioning sessions, sign-in, API authorization, or conversations. The draft may change before version 1.0.
Scan your site and inspect `checks.discovery.personalAgentProtocol`. Its `details.issuerValidation` reports `verified`, `notRequired`, `unverified`, `invalid`, or `notChecked`. Use the audit evidence to diagnose invalid fields, redirects, and issuer/domain mismatches. Network failures and unsupported major versions return `unableToCheck` while retaining available discovery evidence.
26 production-ready skills for AI coding agents. Install with one command, use across Claude Code, Cursor, Kiro, Windsurf, and OpenCode.
Edit English prose to remove repetitive AI phrasing, inflated language, and mechanical structure while preserving meaning and the…
Reescreve textos em português brasileiro para soarem mais humanos e naturais, reduzindo padrões típicos de escrita gerada por IA…
Audit, score, and compare repositories containing portable Agent Plugins against the official Agent Plugins specification. Use…
Audit and improve website readiness for AI agents using the Cloudflare "Is It Agent Ready?" scanner (isitagentready.com). Covers…
Sub-skill de agent-ready-cloudflare: Implement A2A Agent Card
Sub-skill de agent-ready-cloudflare: Implement ACP Discovery Document