AI Skills Hub - Browse 180,000+ Agent Skills

Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.

CLI to manage emails via IMAP/SMTP. Use `himalaya` to list, read, write, reply, forward, search, and organize emails from the terminal. Supports multiple accounts and message composition with MML (MIME Meta Language).

Manage Apple Reminders via the `remindctl` CLI on macOS (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.

Manage Things 3 via the `things` CLI on macOS (add/update projects+todos via URL scheme; read/search/list from the local Things database). Use when a user asks Clawdbot to add a task to Things, list inbox/today/upcoming, search tasks, or inspect projects/areas/tags.

Display HTML content on connected Clawdbot nodes (Mac app, iOS, Android).

Use when writing database access code, creating schemas, or managing transactions with PostgreSQL - enforces transaction safety with TX_ naming, read-write separation, type safety for UUIDs/JSONB, and snake_case conventions to prevent data corruption and type errors

Use when invalid data causes failures deep in execution - validates at every layer data passes through to make bugs structurally impossible rather than temporarily fixed

Use when writing tests for serialization, validation, normalization, or pure functions - provides property catalog, pattern detection, and library reference for property-based testing

Use when planning features and need current API docs, library patterns, or external knowledge; when testing hypotheses about technology choices or claims; when verifying assumptions before design decisions - gathers well-sourced, current information from the internet to inform technical decisions

Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification

Use when creating new skills, editing existing skills, or verifying skills work before deployment - applies TDD to process documentation by testing with subagents before writing, iterating until bulletproof against rationalization

Error modeling and recovery with TaggedError, catchTag(s), mapError, and retry schedules. Use when defining errors or adding resilience.

Core Effect foundations and style for a coding agent. Use when starting an Effect task, choosing operators, or structuring a small pipeline.

Testing patterns with layers, mocks, and deterministic time. Use when preparing testable services and small smoke tests.

Define services, provide layers, compose dependencies, and switch live/test. Use for DI boundaries and app composition.

Concurrency with Effect.all, forEach concurrency, Fiber lifecycle, race and timeouts. Use for parallelizing tasks safely.

Queue and PubSub patterns, background fibers, and graceful shutdown. Use for decoupling producers/consumers.

Connect to a running ComfyUI instance, queue workflows, monitor execution, and retrieve results. Supports both online (REST API) and offline (JSON export) modes. Use when executing ComfyUI workflows or checking server status.

Generate ComfyUI workflow JSON from natural language descriptions. Validates against installed models/nodes before output. Use when building custom ComfyUI workflows from scratch or modifying existing ones.

Getting started guide and performance tips for JADX-AI-MCP. Use when user asks "getting started", "how to use", "first time", "what tools", "where to start", "batch operations", "performance", "large response", "optimize", "slow", "timeout", or needs help choosing between APK and JAR analysis tools.

Search code using Sourcegraph CLI. Use when (re)searching codebases, finding implementation examples, analyzing code patterns

Search the user's Emacs elfeed RSS feed database containing curated feeds from Reddit, blogs, YouTube, GitHub releases, and newsletters. Use when the user asks about articles they've read, mentions RSS feeds or 'something I read', wants to research topics from their curated sources (programming, AI, security, NixOS, Emacs, etc.), or needs to triage unread items.

Fetches and displays Hacker News stories and comments via CLI. Use when the user asks about HN, Hacker News, tech stories, wants to search HN, or wants to read/analyze HN comments.

创建高质量 MCP（Model Context Protocol）服务器的指南，使 LLM 能够通过精心设计的工具与外部服务交互。在构建 MCP 服务器以集成外部 API 或服务时使用，无论是在 Python (FastMCP) 还是 Node/TypeScript (MCP SDK) 中。

Run unit and integration tests for Catalyst-Relay. Use when asked to test, run tests, verify changes, or check if code works.

Add new core business logic functions to Catalyst-Relay. Use when creating pure functions, ADT operations, or library-consumable code.

Extract form field data from PDFs as a first step to filling PDF forms

Run one deterministic Vibe dispatcher step and print the selected prompt body.

IPC timing: ```c // TEST-INT-210 covers REQ-RTOS-IPC-02 CHECK_TRUE(send_cmd(&cmd)); // assert queue receive within 10ms ```

Continuous Vibe loop runner for Codex.

Fault injection step: ```text // REQ-HIL-FAULT-02; TEST-HIL-120 At t=500 ms, drop comms line; expect safe state within 200 ms. ```

Seed corpus entry (hex): ```text // REQ-FUZZ-CORPUS-02 01 00 10 00 AA BB CC DD ```

Deterministic access to the Vibe prompt catalog.

Apply flow (pseudo): ```c // REQ-OTA-APPLY-02; TEST-OTA-06 if (!verify_manifest(m) || !verify_sig(pkg, m->sig)) abort(); if (m->version <= current_version()) abort(); // anti-rollback if (!write_slot(m->slot, pkg, m->len)) abort(); mark_pending(m->version, m->slot); reboot(); ```

Anti-rollback check: ```c // REQ-SBOOT-ROLL-02; TEST-SBOOT-06 if (m->version < stored_version()) { return false; // reject rollback } ```

GATT characteristic with permissions: ```c // REQ-BLE-GATT-02; TEST-BLE-07 const gatt_char_t glucose_meas = { .uuid = GLUCOSE_MEAS_UUID, .props = READ | NOTIFY, .perm = AUTHENTICATED | ENCRYPTED, .max_len = sizeof(glucose_meas_t), }; ```

Terminology mapping table (YAML): ```yaml # REQ-INTOP-TERM-02 device_code: FLOW_RATE loinc: 8310-5 display: Infusion flow rate ```

Reconnect with backoff: ```c // REQ-WIFI-RESIL-02; TEST-WIFI-07 for (int i = 0; i < MAX_RETRIES; i++) { if (wifi_connect()) break; sleep_ms(100 * (1 << i)); // capped elsewhere } ```

Firmware update guard: ```c // REQ-USB-OTA-02; TEST-USB-07 int handle_fw_update(const uint8_t *buf, size_t len) { if (!verify_signature(buf, len)) return -1; return apply_update_atomically(buf, len); } ```

ECC status handling: ```c // REQ-DINT-ECC-02; TEST-DINT-05 if (ecc_corrected) log_event("ECC_CORRECTED"); if (ecc_uncorrectable) { enter_safe_state(); } ```

Bounded queue send with timeout: ```c // REQ-RTOS-IPC-02; TEST-RTOS-05 bool send_cmd(cmd_t *cmd) { return xQueueSend(cmd_queue, cmd, pdMS_TO_TICKS(10)) == pdPASS; } ```

Brownout interrupt: ```c // REQ-PWR-BROWN-02; TEST-PWR-06 void BOD_IRQHandler(void) { clear_bod_irq(); save_state_crc(); enter_safe_shutdown(); } ```

void *pool_alloc(void) { for (size_t i = 0; i < 8; i++) { if (!pool[i].used) { pool[i].used = true; return pool[i].buf; } } return NULL; // handle failure } ```

Static buffer, no malloc: ```c // REQ-C-MEM-02; TEST-C-05 static uint8_t rx_buf[128]; size_t uart_read_safe(uint8_t *out, size_t max_len) { size_t n = uart_read(rx_buf, sizeof(rx_buf)); if (n > max_len) n = max_len; memcpy(out, rx_buf, n); return n; } ```

Bounded buffer view: ```cpp // REQ-CPP-API-02; TEST-CPP-05 struct BufferView { uint8_t* data; size_t len; };

hal_status_t hal_uart_init(uint32_t baud); hal_status_t hal_uart_write(const uint8_t *buf, size_t len, size_t *written); hal_status_t hal_uart_read(uint8_t *buf, size_t len, size_t *read, uint32_t timeout_ms); ```

bool validate_msg(const msg_t *m) { if (m->len > sizeof(m->payload)) return false; return crc16(m->payload, m->len) == m->crc; } ```

typedef state_t (*handler_fn)(void); typedef struct { state_t state; handler_fn fn; } state_entry_t;

Sensor plausibility check: ```c // REQ-FT-SNS-02; TEST-FT-07 bool validate_pressure(float p_kpa) { return (p_kpa >= 0.0f && p_kpa <= 300.0f); } ```

Interface validation snippet: ```c // CLASS C boundary; REQ-IF-01; TEST-IC-12 int handle_cmd(const cmd_t *cmd, size_t len) { if (len != sizeof(cmd_t)) return -1; if (!crc_ok(cmd, len)) return -2; return dispatch(cmd); } ```

Inspect and edit Jupyter notebook (.ipynb) listing, reading, updating, deleting, and inserting cells. Use whenever an agent must analyze or modify notebook cell content (e.g., reviewing markdown, tweaking code) with precise, per-cell control.

Configure scan-mcp MCP server in Claude Code's global configuration. Use when user wants to setup, configure, initialize, enable, or install the scan-mcp MCP server. Runs preflight checks for prerequisites (Node 22+, SANE tools, tiffcp), helps install missing dependencies, and adds server configuration with user-specified INBOX_DIR.

Convert single-file task backlogs (TASKS.md format) to AutoClaude multi-file spec structure. Use when the user wants to (1) convert existing TASKS.md or similar task backlog files to AutoClaude specs, (2) initialize a new project with AutoClaude-compatible task structure, (3) migrate task definitions from simple markdown to the multi-file spec format with requirements.json, context.json, implementation_plan.json. Triggers on phrases like "convert tasks to AutoClaude", "set up AutoClaude specs", "migrate backlog", "create spec from task".

This skill should be used when writing TypeScript code, using "import defer", "--module node20", "tsc --init", working with ArrayBuffer types, or any TypeScript features from version 5.9 onwards.

Capture PNG screenshots of dex TUI commands for debugging UI layout issues. Use when iterating on CLI interface code or debugging rendering problems.

Build and maintain high-quality CLAUDE.md files that maximize Claude Code efficiency. Use when creating a new CLAUDE.md for a project, auditing/improving an existing CLAUDE.md, or when the user mentions "claude.md", "agents.md", "onboarding Claude", or wants to improve their Claude Code workflow.

Automate version bumping, changelog generation, and release preparation. Creates release PR that triggers GitHub Actions workflow for building, testing, and publishing.

Captures screenshots on macOS systems. Can take full screen screenshots or capture specific application windows (e.g., Google Chrome, Safari, Visual Studio Code). Use when the user needs to capture screenshots on macOS, list available windows, or get information about the currently active window.

Captures screenshots on Linux systems. Can take full screen screenshots or capture specific application windows (e.g., Firefox, Chrome, Terminal). Use when the user needs to capture screenshots on Linux, list available windows, or get information about the currently active window.

Debug failed Explorbot interactions. Analyzes Langfuse exports or log files to find why tests failed and suggests Knowledge fixes.

This skill should be used when the user asks to "add news", "post news", "add an X post to news", "add tweet to news", "add announcement", "add curated link", "post to news feed", "share on news", "add article to news", or mentions adding content to the dcbuilder.dev news section. Provides workflows for adding announcements (portfolio company posts with logos) and curated links (external content) with optional R2 image upload.

Use when the user asks to "add job", "post job", "create job listing", "add a position", "post a role", "add job opening", "list a job", "add candidate to jobs board", or mentions adding someone looking for work to the jobs section.

Use when each language SDK lives in a separate repository. Covers cross-repo workflow dispatch, PR status reporting, PR reconciliation on merge/close. Triggers on "multi-repo SDK", "separate SDK repositories", "cross-repo workflows", "SDK PR synchronization", "spec repo triggers SDK repos".

Use when SDK generation failed or seeing errors. Triggers on "generation failed", "speakeasy run failed", "SDK build error", "workflow failed", "Step Failed", "why did generation fail"

Expert localization decisions for iOS/tvOS: when runtime language switching is needed vs system handling, pluralization rule complexity by language, RTL layout strategies, and string key architecture. Use when internationalizing apps, handling RTL languages, or debugging localization issues. Trigger keywords: localization, i18n, l10n, NSLocalizedString, Localizable.strings, stringsdict, plurals, RTL, Arabic, Hebrew, SwiftGen, language switching

Expert Swift concurrency decisions: async let vs TaskGroup selection, actor isolation boundaries, @MainActor placement strategies, Sendable conformance judgment calls, and structured vs unstructured task trade-offs. Use when designing concurrent code, debugging data races, or choosing between concurrency patterns. Trigger keywords: async, await, actor, Task, TaskGroup, @MainActor, Sendable, concurrency, data race, isolation, structured concurrency, continuation

Expert decision trees for Solargraph, Sorbet, and Rubocop validation in Rails. Use when: (1) Choosing which quality tool for a task, (2) Debugging type errors or lint failures, (3) Setting up validation pipelines, (4) Deciding strictness levels. Trigger keywords: quality gates, validation, solargraph, sorbet, rubocop, type checking, linting, code quality, static analysis, type safety, srb tc

通过openspec规范驱动的方法创建结构化的变更提案与规范差异。用于规划功能、创建提案、编写规范、引入新能力或启动开发流程。触发词包括 "openspec提案", "规划", "创建提案", "规划变更", "规范功能", "新功能", "新特性", "新需求", "添加功能规划", "设计规范"。

执行实施规划工作流程，使用计划模板生成设计工件。触发词包括："speckit计划"。

Create execution plans (PLAN.md files) for a roadmap phase. Spawns a planner agent to decompose phase goals into executable plans with tasks, dependencies, and wave structure. Includes plan verification loop. Use after /create-roadmap to plan a specific phase before execution. Supports --all to plan every unplanned phase sequentially. Supports gap closure mode (--gaps) for fixing verification failures.

Check project progress and route to next action. Analyzes .planning/ files to show current position, recent work, key decisions, and intelligently routes to the appropriate next step (/new-project, /create-roadmap, /plan-phase, /execute-phase, etc.). Use when user asks "where am I?", "what's next?", returns to project after time away, or completes a phase and needs direction.

Manage and fix Claude Code plugin marketplaces -- audit for issues, register/remove/create plugins, update entries, sync versions, configure distribution, and maintain marketplace.json manifests. Use when working with marketplace catalogs, plugin entries, team distribution settings, or fixing marketplace inconsistencies.

Trade on Perpl DEX - view markets, manage positions, execute trades

Show current Specwright workflow status. Displays active epic, task progress, gate results, and learning queue size.

Cloud cost forecasting and budget planning. Project future costs based on trends, model growth scenarios, and provide budget recommendations with confidence ranges. Use when: "forecast costs", "budget planning", "project spending", "future costs", "capacity planning costs", "cost projections"

Guide for writing E2E test scenarios in the wicked-scenarios format. Covers scenario structure, frontmatter fields, step format, and best practices. Use when: - Creating new test scenarios - Understanding the scenario format - Choosing the right CLI tool for a test category

Aggregate customer feedback from discovered sources across support, surveys, social, and direct channels. Use when you need to gather customer voice data to inform product decisions or understand customer sentiment.

WCAG 2.1 Level AA accessibility audit and remediation guidance. Keyboard navigation, screen readers, semantic HTML, ARIA patterns. Use when: "accessibility", "a11y", "WCAG", "keyboard", "screen reader", "semantic HTML", "ARIA", "inclusive design", "disability"

GitLab CLI (glab) utilities - pipeline debugging, MR management, release automation. Use when working with GitLab via CLI for anything beyond basic git commands.

Write secure, optimized GitHub Actions workflows. Use when creating CI/CD pipelines, automating deployments, or debugging workflow issues. Security-first approach with performance optimization.

This skill should be used when the user asks to "track tasks", "create a task", "add a todo", "manage tasks", "show my tasks", "what should I work on", "move task to done", "update task status", or mentions persistent task tracking, kanban boards, or project management. Provides guidance for using wicked-kanban for persistent task management.

Maintain a single-file interactive industry map (`index.html`) with evidence-tiered relationships, clean visual hierarchy, strong logo quality, USD-normalized market text, and publish-ready repository hygiene.

安全隔离的 Docker 沙盒代码执行服务。支持 Python/Shell/Bash 多语言动态执行，内置超时与资源限制。提供信任模式用于服务间代码融合调用。

按优先级并发实现 Project 中所有 Issue。

Project 级别批量 PR 审查与合并。

同步 Issue 到 GitHub Projects 看板。

Run PostgreSQL queries and meta-commands via CLI

Manage tmux sessions for interactive background processes

GitHub PR utilities for code review workflows

Update feature progress log and check off completed tasks. Use when user completes implementation tasks, makes commits, or indicates work is done. ASKS BEFORE MODIFYING files. Updates plan.md progress log section and implementation step checkboxes.

Interactive workflow for adding items to the JSON backlog. Use when user wants to add a new feature, track an idea, capture requirements, or mentions something that should be in the backlog.

Use when releasing new versions for Debian projects with optional linglong.yaml support, updating debian/changelog, incrementing patch versions automatically, and creating git commits

Use when user requests translating Qt project localization files (TS files), automating translation workflows, or setting up multilingual support for Qt applications. This skill uses parallel processing with ThreadPoolExecutor to translate TS (Translation Source) files efficiently.

开发日志存档工具。将日常工作中的关键节点（调试、方案设计、故障排查）结构化记录到 Markdown 文件。支持全局/项目本地双模式存储，具备智能防重功能，便于后续生成周报。

Comprehensive ahooks React hooks library specialist. Expert in all 76+ ahooks hooks including state management, effects, data fetching, performance optimization, DOM utilities, and advanced patterns. Use when working with ahooks library, need React hooks utilities or want to learn best practices.

Creates engagement agendas from planning transcripts or notes. Extracts structured agenda data into JSON format, then renders DOCX using templates. Use when building agendas, processing planning calls, or when user mentions transcripts, Discovery, Architecture Review, POC, Executive Briefing, Deep Dive sessions, or agenda generation.

Generates engagement task timelines (T-28 to T+3) with business day calculations excluding weekends. Outputs Planner-ready CSV files. Use when generating tasks, creating timelines, or when user mentions task lists, Planner import, or business day calculations.

Sync Discord messages using USER TOKEN. Supports servers AND DMs. Use when routed here by community-agent:discord-sync or user explicitly requests user token sync.

Read and search synced Telegram messages. Use when user asks about Telegram conversations, wants to see messages, or search for specific content.

Initialize Telegram configuration. Use when user wants to set up, configure, or connect their Telegram account for the first time.