Agent Skills (Open Standard)
SDK / ToolThe open Agent Skills standard — SKILL.md folders any agent can load, governed by the Agentic AI Foundation since 2026-10-22.
- official
- spec
- agent-skills
- standard
A local Manifest V3 Chrome extension that reads your organizations' KYC status from a logged-in claude.ai tab — it never touches sessionKey or cookies, and every request stays on claude.ai
claude-kyc-checker (Chinese title: Claude KYC 检查器) is a local Chrome extension for checking the KYC (know-your-customer) status of every organization on your claude.ai account. It calls two endpoints from inside the logged-in claude.ai tab via chrome.scripting.executeScript with credentials included, so the browser itself carries the login state: GET /api/organizations for the organization list, then GET /api/organizations/<uuid>/kyc_status per organization. The popup lists each organization's name, shortened UUID, plan capabilities (claude_pro, claude_max, raven — as returned by the API), KYC status, a classification of the verification link, check_kind and a timestamp rendered in Asia/Shanghai. Status values and their documented meanings: not_required (upgrade needs no KYC), pending (KYC required or under review), verified — the older API said approved (KYC passed), denied (KYC rejected, cannot upgrade). When the response carries a vendor_inquiry_url the extension classifies it: Persona subscription KYC (links under inquiry.withpersona.com) is treated as routine subscription verification, while links to portal.anthropic.com are flagged as portal KYC, which public tutorials associate with account-ban risk — the extension warns and recommends against proceeding — and unknown links are flagged for caution. A copy button puts the raw JSON of the last query on the clipboard. Privacy posture: no server, no analytics, permissions limited to activeTab and scripting plus a claude.ai host permission, no cookies API, no credentials stored or transmitted anywhere else. MIT.
Use when a claude.ai upgrade or plan change is stuck and you need to see whether KYC is the blocker — which organizations are pending, verified or denied, and which kind of verification link the account was handed — from a local tool that stays inside the claude.ai origin and carries no third-party hop.
Load unpacked (Manifest V3): 1. Download the repo (Code -> Download ZIP) and use its folder directly 2. Open chrome://extensions 3. Toggle Developer mode (top right) 4. Click Load unpacked and select the folder containing manifest.json Then open and log in to claude.ai, click the extension icon, and Refresh. Copy Results copies the query JSON. Dev checks: node --check classify.js; node --check popup.js; node --check test-classify.js; node test-classify.js — the tests cover sample JSON for all statuses and both link classes, and never touch a real account.
It talks to two undocumented internal claude.ai web endpoints, so response shapes and the status vocabulary can change without notice; the README itself frames the status meanings as common values, not a spec. The portal-KYC ban-risk claim is attributed to public tutorials, and the extension's warning is advisory, not something Anthropic documents. Scope: it reads KYC status only — it is not a plan manager or an upgrade tool, and the README asks you not to do automated repeated probing of real accounts beyond the extension's own read-only requests. Timestamps render in Asia/Shanghai regardless of your locale. Unofficial, not affiliated with Anthropic. MIT license. 24 stars as of writing (2026-11-16).
Repository
VallierDev/claude-kyc-checkerCharted
The open Agent Skills standard — SKILL.md folders any agent can load, governed by the Agentic AI Foundation since 2026-10-22.
A local launcher that points Claude Code and seven other coding harnesses at any OpenAI-, Anthropic- or Responses-compatible endpoint — without editing a single config file.
Answers macOS Spotlight and Siri prompts with your signed-in Claude Code account — an experimental macOS 27 model-delegation provider.