Browse the metahub catalog

Use when the cardputer MCP tools (notify, ask, confirm, show, progress) are available in the session, or the user mentions their Cardputer / handheld / "buzz me" / "page me". Governs runtime etiquette for the pocket device — mandates a physical confirm gesture before irreversible operations (passing the real action diff as details), buzzes the device on completion of long tasks, asks quick questions when blocked and the user is away from the keyboard, shows ambient status for long work, and formats all device output for the 240×135 LCD. This is the behavioral counterpart to the cardputer MCP server: the server is the hands, this skill is the manners. Trigger even when the user doesn't name the skill — if the `cardputer` MCP tools are registered, these rules are in force.

Domscribe is a pixel-to-code development tool that maps running UI elements to their exact source locations, capturing runtime context (props, state, DOM) for handoff to coding agents via MCP.

Modular iOS development automation suite. 8 workflow-specific MCP servers (600-3500 tokens) with 23 tools across Xcode, Simulator, and IDB. Enable only what you need!

Use when the user says: 'ijfw status', 'show me the system state', 'what is ijfw doing', or to see the at-a-glance routing / memory / activity banner.

Diagnose IJFW integration health per platform. Trigger: 'doctor', 'check setup', /doctor

AI efficiency framework. Smarter output, intelligent routing, persistent memory, context discipline, project workflows. One install. Zero config. It just fucking works.

Cross-platform persistent memory server for IJFW. 14 MCP tools (memory + admin/update + brain). Works with 15 platforms: 14 via MCP (Claude Code, Codex, Gemini CLI, Cursor, Windsurf, Copilot, Hermes, Wayland, OpenCode, QwenCode, Cline, KimiCode, OpenClaw, Antigravity) plus Aider via the rules-only tier.

One-command installer for IJFW -- the AI efficiency layer. One install, every AI coding agent, zero config.

>

How to cut a portainer-mcp release that bumps the embedded Portainer OpenAPI spec to a new Portainer minor (e.g. 2.41.x → 2.42.x). Walks through finding the upstream patch target, regenerating the spec, recounting ops/tags per profile, refreshing the orphan-tag inventory, bumping version pins across the README and distribution docs, promoting CHANGELOG entries, committing, pushing, merging to main, and tagging — in that order. Trigger this skill whenever the user mentions cutting a release, bumping the Portainer version target, supporting a new Portainer minor, regenerating the spec, running `make specs` for a new minor, or any phrasing like "let's update for Portainer X.Y" — the spec regen is only the first step of a multi-file release and the model should consult this skill before touching anything.

How to drive the Portainer MCP server's tools correctly — both reading and mutating. Reading: project responses with `select` (JMESPath), where the heavy fields live (snapshots, status blocks, managed fields), how to handle non-JSON Docker/K8s proxy endpoints (container and pod logs, stats, exec), and how to interpret results that are easy to misread (e.g. an edge environment's health comes from its heartbeat, not its `Status` field; typed K8s tools use different field names than the raw proxy). Mutating: deploying, scaling, restarting, and deleting Portainer-managed resources — where success payloads are empty and must be verified out-of-band, and where cleanup (orphaned volumes) and recovery (name-based vs id-based calls) have gotchas. Trigger this whenever you're about to call any Portainer MCP tool — including `docker_proxy`, `kubernetes_proxy`, `EndpointList`, `GetAllKubernetes*`, `StackList`, `StackCreateKubernetes*`, `StackDelete*`, `CreateKubernetes*`, `snapshot*`, `Helm*`, or any other `mcp__portainer__*` tool — and whenever the user asks you to inspect, deploy, scale, restart, or remove Docker containers/images/stacks/networks, Kubernetes resources, or Helm releases managed by Portainer. Use it even if the user doesn't mention Portainer by name, as long as the working answer requires one of these tools.

Update library documentation in .claude/docs/libraries/ with latest information from web search.

Implement features using Test-Driven Development (TDD) with Red-Green-Refactor cycle.

Simplify and refactor code while preserving functionality and library constraints.

Research a library and create comprehensive documentation in .claude/docs/libraries/.

Create a detailed implementation plan for a feature or task. Use when user wants to plan before coding.

Analyze project structure, write the thick requirements doc (.claude/docs/DESIGN.md), populate the thin "Repository Identity" pointer in CLAUDE.md (Zone B), and mirror identity into AGENTS.md.

PROACTIVELY track and document project design decisions without being asked. Activate automatically when detecting architecture discussions, implementation decisions, pattern choices, library selections, or any technical decisions. Also use when user explicitly says "record this", "what's our design status", or equivalent. Do NOT wait for user to ask - record important decisions immediately.

Independent adversarial honesty reviewer for a DRAFT output: checks claims-vs-evidence (no fabrication, no overreach, no invented IDs) before a human approves. FULL mode reads an agentic-sop-kit run dir's trace; DEGRADED mode reviews any plain document with user-supplied inputs. Deterministic honesty checks in code + a capped ADVISORY adversarial LLM pass. The report is itself a DRAFT — a human owns the verdict; it never auto-approves or auto-rejects. An optional project-scoped Stop-hook can enforce that a human acknowledged a review before the session ends. Not a source-code reviewer.

Claude Code plugin linter + scaffolder. `lint` validates a plugin (or a whole marketplace) against a house grammar — a strict superset of the usual manifest/frontmatter checks plus stdlib-only, hook-protocol, and test-harness invariants; `scaffold` generates a grammar-conformant plugin skeleton that passes the linter by construction. Pure stdlib; self-hosting; runs standalone on any plugin or snaps onto this marketplace.

Deterministic ALCOA+ data-integrity linter for a record/dataset: checks Attributable / Contemporaneous / Complete / Accurate / Consistent in code, and surfaces the human-judgment slice (authorization, gap justification) as a 【待補】 checklist it never auto-concludes. DEGRADED mode lints any CSV/JSON record against a declared contract; FULL mode snaps onto an agentic-sop-kit run dir. The report is a DRAFT — a human owns the verdict. Domain-neutral engine, GMP-flavored ruleset. Pure stdlib.

Convert human SOPs into deterministic agentic workflows. Bundles the six-rung-ladder + agentic-sop + agentic-workflow-audit skills, the /sop-flow command, a project-scoped Stop-hook regression gate, and the portable agentic-sop-kit.

Design and verify data contracts, schema changes, migrations, rollbacks, backfills, compatibility windows, tenant isolation, and API-to-database field mapping. Use when changing database schema, event payloads, persisted documents, analytics tables, multi-tenant data boundaries, or any user-visible data contract where silent fallback or undeclared fields would be dangerous.

Use this skill to diagnose CSS and frontend layout issues such as positioning, overflow clipping, Tailwind class conflicts, z-index stacking, and React rendering visibility problems.

End-to-end open source contribution workflow: from scanning issues to submitting PRs. Use this skill whenever the user wants to contribute to an open source project, find issues to fix, submit a pull request, fork a repo to contribute, fix a GitHub issue, or mentions 'open source contribution'. Also trigger when they provide a GitHub repo URL and ask about contributing, say things like 'help me submit a PR', 'find good first issues', 'I want to contribute to X', or mention fixing bugs in someone else's project.

Use when a contributor wants to move beyond simple bug fixes into architectural improvements, technical debt discovery, design proposals, or module ownership opportunities.

Design, audit, and verify configuration, environment separation, secrets, BYOK flows, key rotation, config schema validation, and drift checks across local, dev, staging, and production. Use when adding env vars, changing runtime config, handling API keys or user-provided keys, diagnosing config drift, or preparing deploy/release configuration.

Complete testing strategy covering TDD workflow, test pyramid, unit/integration/E2E/property testing, framework best practices (Jest, Vitest, pytest), mock strategies, and CI integration. Use when writing tests, reviewing test quality, or establishing testing standards.

Use when you want Codex to review its own recent history (last N days or specific period) and improve its behavior. Produces minimal, high-signal updates to AGENTS.md and tiny reusable skills. The goal is long-term fluency — Codex gradually becomes better at your specific style, constraints, and workflows.

Automatically diagnose excessive Codex local SQLite diagnostic log writes and give a concrete fix plan. Use when a user asks whether Codex is writing too much to disk, mentions logs_2.sqlite, logs_2.sqlite-wal, block_log_inserts, SSD/TBW wear from Codex logs, or wants Codex log write issues checked, explained, stopped, cleaned up, verified, or restored.

Use when Codex feels slow, bloated, or heavy after heavy use. Provides safe session hygiene, archive strategy, and handoff discipline to keep daily Codex usage responsive and low-friction. Always pairs with comprehensive handoffs before archiving active work.

Use when you want a second-opinion review via Codex CLI, cross-verification after another agent implements changes, debugging help, or alternative implementation proposals. Requires Codex CLI to be installed and authenticated.

全面代码库审计 — 自适应并行深度分析（前后端契约、数据完整性、异常处理/安全、架构/技术债、配置/缓存），结构化 findings + 对抗验证 + 基线对比，输出按严重程度排序的统一报告和修复路线图。支持 quick 快速体检模式。Use when user asks to audit, analyze, or review an entire codebase for design issues, find hidden bugs, check architecture health, or asks '全面审查', '代码库审计', '分析设计问题', 'audit codebase', 'health check', '有哪些问题', '快速体检'. Also trigger when user asks to find silent degradation, data flow breakpoints, type mismatches between frontend and backend, or wants to understand technical debt across a project.

在 Linux VPS 上部署 CLIProxyAPI + NewAPI 组合栈，把 Codex/Claude/Gemini/Qwen 等订阅账号包装成可计费的 OpenAI 兼容 API。负责 NewAPI Docker 部署、容器→宿主桥接、模型计费倍率（ModelRatio/CacheRatio/CompletionRatio）、SQLite 直写额度、CLIProxyAPI 408 冷却补丁、多账号 OAuth 凭据热加载、双路径验证。当用户说"部署 cliproxy + newapi"、"加 OpenAI 计费层"、"配置 newapi 渠道接 cliproxy"、"newapi 价格不对"、"加新的 codex/claude 账号到现有部署"、"172.17.0.1 容器网络"、"408 冷却放大故障"时触发。

在 Linux VPS 上部署 CLIProxyAPI（router-for-me/CLIProxyAPI）并暴露 OpenAI 兼容 API。覆盖 Go 环境、源码拉取、最小配置、OAuth 登录（远端 / 本地同步两种路径）、UFW 放行、密钥轮换、健康校验、环境变量落盘。当用户说“部署 cliproxy / 把 codex/claude/gemini 账号暴露成 API / router-for-me 上线 VPS”时触发。

查看指定进程的代理线路。通过 Mihomo API 查询当前活跃连接，显示进程匹配的规则和代理链路。用于确认某个进程（如 claude、chrome）走的是哪条订阅线路

Clash Verge 诊断与配置管理。网络诊断、配置 AI 工具路由规则（Claude/Codex/Cursor/Warp）、查看/克隆订阅配置、切换配置文件。当用户遇到网络问题、要配置代理规则、切换订阅、查看 Clash 配置时使用

使用 auto-run-agent 对目标项目进行自动化分析、评估、设计和优化。完整流程包括：探索项目现状 → 识别问题 → 设计任务列表 → 创建 runner 环境 → 启动自主执行。适用于任何需要批量优化的项目。

OAuth 2.1 + JWT authentication security best practices. Use when implementing auth, API authorization, token management. Follows RFC 9700 (2025).

Ask Grok or Gemini via opencli (browser-session based, no API cost). Use when the user says "问 grok", "问 gemini", "ask grok", "ask gemini", "用 grok 查", "grok 怎么看", "让 gemini 分析", or wants a second opinion from Grok/Gemini without paying API tokens. Routes to opencli which drives the already-logged-in Chrome session.

Design architecture foundations before implementation. Use when asked to design or refactor architecture, choose Rust/Go crate, package, module, runtime, workflow, or service boundaries, compare mature project architecture, prevent stacked one-off PRs, audit migration debt in existing repos, or produce a spec before coding.

End-to-end app feature inventory and user-story testing workflow. Use when the user asks to audit every feature, create user stories or expected behavior from code, maintain a single canonical spreadsheet/tracker, test each user behavior, document errors, fix logistical or UX issues, and retest after fixes.

Mobile app UI design expert for iOS and Android. Use when designing app interfaces, creating design systems, ensuring accessibility, or following platform guidelines. Covers Material Design 3, Human Interface Guidelines, color theory, typography, and 2025 trends.

REST/GraphQL/gRPC API design best practices. Use when designing APIs, defining contracts, handling versioning. Covers OpenAPI 3.2, GraphQL Federation, gRPC streaming.

Generate or update repository-specific AGENTS.md instruction files from real repo evidence. Use when asked to create, design, scaffold, split, or improve root or scoped AGENTS.md files for Codex/Claude/agent workflows, especially when a repo needs directory-specific rules, validation commands, generated-file boundaries, or a short agent onboarding router.

Audit AND optimize a CLAUDE.md / AGENTS.md instruction file — score it against the five high-leverage patterns, flag anti-patterns, then apply approved fixes in place. Use when the user says 优化 CLAUDE.md / 优化 AGENTS.md / optimize my agent doc / 帮我改 claudemd, or after an audit when they want the fixes applied (not just reported).

AI Agent Team Operating System for Claude Code with 155 MCP tools (incl. 40+ ecosystem research tools), 25 agent templates, 14 hooks / 12 lifecycle events. Persistent team management, structured meetings, task wall with pipeline workflows, company loop engine, real-time React dashboard, and Ecosystem Research Platform v2 (progressive 4-stage deep-review funnel: shallow auto-summary → on-demand architecture → debate-based finalist evaluation → reference/integrate marking, with project-customizable thresholds, append-only history snapshots, and Failed self-learning).

Cloud-backed persistent memory powered by Deeplake — read, write, and share memory across Claude Code sessions and agents