All authors
Claude Skills by youlianvr
github.com/youlianvr269 skills0 installs0 views
- Skill MetaMeta-skills about the skills layer itself: creating new skills (scaffold + evals), validating and scoring skill quality, the authoring spec (frontmatter, folder structure, the unified router shape, checklists), installing community skills, the local MCP-ecosystem map, and plugin bundles. Use when the task is about skills themselves — "create a skill", "test this skill", "install a community skill", "audit skill quality".Votes: 0GitHub stars: 15
- Plugin CreatorPackage several skills, and optionally an MCP server, as one versioned bundle with a manifest and an explicit trust review. Use when a set of skills is meant to travel together, or when a client needs a single installable unit instead of loose folders.Votes: 0GitHub stars: 15
- Skill AuthoringUse when creating or editing any skill in `skills/` — frontmatter rules, folder structure, the unified router shape, script bundling, and the quality checklist.Votes: 0GitHub stars: 15
- Skill CreatorCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, optimize a skill's description for better triggering accuracy, or validate a skill's structure against a review checklist (write-a-skill doctrine merged 2026-08-05).Votes: 0GitHub stars: 15
- Skill HubMap of the local MCP ecosystem: BrowseMCP (86 browsing tools that act \"as a human\"), SearchMCP (web research in one call) and the mcp-repair skill (repairing MCP connections). The learned → done chain: research → navigate → extract_*. Plus the map of harnesses (skill and rules paths), the skill inventory with canonical sources, the farm lifecycle (skills_install.py, --inventory audit), and the MCP tool-prefix table per client. Use when you need to know which server to use, what is visible i...Votes: 0GitHub stars: 15
- Skill InstallerInstall, update, trust or inspect a skill that comes from outside this workspace. Use when the user asks which skills are available, wants a community skill installed, or asks whether a downloaded skill can be trusted.Votes: 0GitHub stars: 15
- Skill TesterValidate, test, and score the quality of skills within the claude-skills ecosystem. Comprehensive meta-skill: structure validation, Python script testing (syntax + imports + runtime + output format), multi-dimensional quality scoring with letter grades and tier classification (BASIC/STANDARD/POWERFUL). Use when authoring a new skill, auditing existing skills for tier promotion, setting up pre-commit hooks for skill quality, or integrating skill QA into CI.Votes: 0GitHub stars: 15
- TelegramTelegram communications, bot operations, formatted reports, channel/inbox triage, and bot hosting strategies. One router over specialized parts: telegram-report, telegram-rich-messages, telegram-bot-hosting-triage, telegram-serverless, content-delivery-format, and tg-check.Votes: 0GitHub stars: 15
- Content Delivery FormatUse when the user wants to: deliver results in different formats (text/rich/PDF/HTML), split a long response into messages, add fallback for a new delivery format, implement a new API method not supported by the library, or gradually enable a new format. Covers: rendering by content type, line-aware split, try-new→fallback-old, native API call, incremental rollout, payload validation, rich delivery in history. Do not use for promises/copy (product-promise-contract), navigation (ux-navigation-...Votes: 0GitHub stars: 15
- Telegram Bot Hosting TriageChoose a practical hosting and API strategy for a Telegram LLM bot, separating bot runtime, model API, persistence, and webhook/polling constraints. Use when someone asks where to host a Telegram AI bot, wants a free 24/7 deployment, has no API key or infrastructure, or is deciding whether to build, self-host, or use an existing public bot.Votes: 0GitHub stars: 15
- Telegram ReportTelegram reporting protocol, kept as a historical reference: how a task report to the owner was shaped — success, error, question — with threading on the message being answered. Inactive by default: do not send automatic reports. Use when the user asks how such reporting worked or wants it revived.Votes: 0GitHub stars: 15
- Telegram Rich MessagesSend Telegram Rich Messages via Bot API 10.x sendRichMessage: collapsible expandable blockquotes, details blocks, tables, checklists, code blocks, spoilers, date widgets, dividers, footers — all in ONE message. Contains LIVE-VERIFIED JSON shapes (many differ from the official docs!), a ready Python sender, a probe script, and the troubleshooting table of API error messages. Use when posting changelogs, reports, digests, or any formatted content to Telegram chats/threads/topics. Triggers: rich...Votes: 0GitHub stars: 15
- Telegram ServerlessTelegram Serverless (tgcloud, cloud.telegram.org) bot hosting: project layout, the push/migrate/run CLI loop, what the V8 isolate actually provides, the invocation time ceiling, where module secrets have to live, and the cutover from a polling container. Use when hosting or moving a bot onto Telegram's own platform, when a handler dies at about thirty seconds or answers nothing at all, when a module needs a provider key, when a webhook and a container fight over getUpdates (409), or when deci...Votes: 0GitHub stars: 15
- Tg CheckTelegram inbox analysis: read per-channel folders in inbox/, extract findings (API keys, tools, services, freebies, attachments) and report to the user. Triggered by "check tg", "what's new in tg". This is the READ side: the fetcher is not run, only already-downloaded messages are analyzed.Votes: 0GitHub stars: 15
- Triage RouteTriage and routing of a reverse-engineering task: by signal (what the target is, what material exists, what result is wanted) choose which reverse-engineering skill to load first. Use ALWAYS at the start of any RE task — before reading other skills. Not for ordinary coding.Votes: 0GitHub stars: 15
- VpnVPN setup, repair, and verification: GUI client one-button setup (v2RayTun), system-wide stack bootstrap and self-reflection (broken wrappers, RU-zone whitelists, TUN mode, autorun), sing-box TUN tunnel reference on Fedora/Linux, and Android VPN/proxy detection. One contract; the parts under parts/ carry the procedures.Votes: 0GitHub stars: 15
- Singbox TunFedora/Linux reference for a system TUN tunnel through sing-box. Installation, systemd, CAP_NET_ADMIN, routing and DNS changes are mutation procedures requiring explicit authorization.Votes: 0GitHub stars: 15
- Vpn Gui SetupSet up VPN via GUI client (v2RayTun) to a single button: installation, subscription import via deeplink without manual clicks, disabling "whitelists" at config level, TUN mode, one-button control. Self-reflection: GUI buttons are unreliable, Flutter doesn't support PrintWindow, deeplink is a deterministic path, config is JSON in shared_preferences. No real IP/country/provider. Triggers: "set up VPN", "v2raytun", "single VPN button", "disable whitelist", "vpn gui".Votes: 0GitHub stars: 15
- Vpn Stack FixSelf-reflection and bootstrap of a system-wide VPN stack: client wrapper broken → cores launch directly; "whitelists" RU-zone in subscription → IP-checker sites show local IP; TUN mode for entire PC; autorun via Task Scheduler; scripts for on/off/server switch. No real IP, country or provider — only generalized formulations. Triggers: "fix VPN", "vpn doesn't work", "happ doesn't work", "2ip shows my IP", "raise vpn-stack", "vpn-on", "how to fix vpn".Votes: 0GitHub stars: 15