发现 Skills

Automatically report fail2ban-banned IPs to AbuseIPDB and notify via Telegram. Use this for server security monitoring, attacker reporting, or checking banned IPs. Watches fail2ban for new bans, reports them to AbuseIPDB, and sends alerts.

飞书视频会议：查询会议记录、获取会议纪要产物（总结、待办、章节、逐字稿）。1. 查询已经结束的会议数量或详情时使用本技能(如历史日期｜ 昨天 | 上周 | 今天已经开过的会议等场景)，查询未开始的会议日程使用 lark-calendar 技能。2. 支持通过关键词、时间范围、组织者、参与者、会议室等筛选条件搜索会议记录。3. 获取或整理会议纪要时使用本技能。

Query and manage Linear issues. Triggers: linear issue, team metrics, WIP.

JikiME-ADK's foundational principles - TRUST 5, SPEC-First DDD, delegation patterns, token optimization, progressive disclosure, modular architecture, agent catalog, command reference, and execution rules for building AI-powered development workflows

JavaScript ES2024+ development specialist covering Node.js 22 LTS, Bun 1.x (serve, SQLite, S3, shell, test), Deno 2.x, testing (Vitest, Jest), linting (ESLint 9, Biome), and backend frameworks (Express, Fastify, Hono). Use when developing JavaScript APIs, web applications, or Node.js projects.

Vercel edge deployment specialist covering Edge Functions, Next.js optimization, preview deployments, ISR, and storage solutions. Use when deploying Next.js applications, implementing edge computing, or configuring Vercel platform features.

Comprehensive codebase architecture mapping with AST analysis and dependency visualization

Eval-Driven Development (EDD) framework - treat evals as unit tests for AI development

For ordered processing: A* search, Dijkstra, event simulation, task scheduling. Efficient min/max extraction with heap-based queue.

For boundary conditions: empty collections, zero values, recursive base cases, null checks, prevent crashes at edges.

For combinatorial iteration: permutations, combinations, cartesian products, without storing all results in memory.

For domain-specific languages: operator overloading, make Python look like math/domain notation, expression builders.

Manage parallel Claude Code workstreams using git worktrees. Use when: splitting large tasks across multiple workers, coordinating parallel development, monitoring worker progress, integrating completed work, analyzing work item documents (code reviews, issue lists). Triggers: parallel, orchestrator, worktrees, workers, coordinate, integrate, split work, multiple sessions, code review, work items.

For long functions: break into smaller pieces, extract helper functions, reduce nesting, improve testability and readability.

For example-driven development: test cases as specifications, input/output pairs, documentation through examples.

For counting and frequency: histograms, most common elements, vote tallying, neighbor counting, word frequencies, distribution analysis.

For grouping and automatic initialization: adjacency lists, grouping by key, multimaps, and nested structures without explicit KeyError handling.

For functional operations: named functions for operators, cleaner than lambdas for arithmetic, building interpreter environments.

Use this skill when the user needs research-group mentoring, lab-management, or lab-culture help that turns expectations into concrete systems, documents, routines, scripts, or escalation paths. Trigger on lab handbooks, onboarding, mentoring compacts, IDP meetings, feedback scripts, authorship or credit norms, conflict diagnosis, performance concerns, PI/lab leadership systems, power dynamics, belonging, work-life boundaries, and culture repair in scientific labs. Do not trigger for personal career decisions, research-project strategy, manuscript/grant work, or legal/clinical determinations.

Jailbreak API-served LLMs using G0DM0D3 techniques — Parseltongue input obfuscation (33 techniques), GODMODE CLASSIC system prompt templates, ULTRAPLINIAN multi-model racing, encoding escalation, and Hermes-native prefill/system prompt integration. Use when a user wants to bypass safety filters, jailbreak a model, or red-team model robustness.

Set up and run HeartMuLa, the open-source music generation model family (Suno-like). Generates full songs from lyrics + tags with multilingual support.

llama.cpp local GGUF inference + HF Hub model discovery.

Drive the macOS desktop in the background — screenshots, mouse, keyboard, scroll, drag — without stealing the user's cursor, keyboard focus, or Space. Works with any tool-capable model. Load this skill whenever the `computer_use` tool is available.

Monitor system health via Glances REST API (CPU, memory, disk, network, sensors, containers, processes). Use when the user asks to "check glances", "system stats", "CPU usage", "memory usage", "glances status", "server health via glances", "disk space", "network traffic", "container stats", "process list", "sensor temperatures", or mentions Glances/system monitoring.