notebooklm
Automate Google NotebookLM - create notebooks, add sources, generate podcasts/videos/quizzes, download artifacts. Supports browser-based content extraction for anti-crawler platforms (Twitter/X, LinkedIn, etc.)...
Découvrez les skills development productivity tools, comparez leurs dépôts et consultez leurs instructions.
Automate Google NotebookLM - create notebooks, add sources, generate podcasts/videos/quizzes, download artifacts. Supports browser-based content extraction for anti-crawler platforms (Twitter/X, LinkedIn, etc.)...
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or to...
Emulates the AI Coding Toolkit's Claude Code command /progress (show progress through EXECUTION_PLAN.md by phase: completed vs remaining acceptance criteria, last completed task, next task). Triggers on "/progr...
Emulates the AI Coding Toolkit's Claude Code command /phase-checkpoint (run checkpoint criteria: tests, typecheck, lint, security scan, coverage; list manual verification steps; update .claude/phase-state.json...
Emulates the AI Coding Toolkit's Claude Code command /fresh-start (orient to project structure, validate directory, load AGENTS.md + specs + EXECUTION_PLAN.md, optionally set up git/feature branches). Triggers...
Emulates the AI Coding Toolkit's Claude Code command /phase-prep (checks prerequisites before starting a phase: pre-phase setup, prior phase completion, git status, tool availability). Triggers on "/phase-prep...
Emulates the AI Coding Toolkit's Claude Code command /verify-task (verify a specific task from EXECUTION_PLAN.md against acceptance criteria, enforce TDD expectations, fix failures with a limited retry loop)....
Emulates the AI Coding Toolkit's Claude Code command /populate-state (generate .claude/phase-state.json from EXECUTION_PLAN.md + git history to track phase/task status). Triggers on "/populate-state" or "popula...
Emulates the AI Coding Toolkit's Claude Code command /list-todos (analyze TODOS.md, prioritize items, and produce implementation guidance without guessing when requirements are unclear). Triggers on "/list-todo...
Analyze completed development work to identify automation and systematization opportunities. Use after finishing features, fixing bugs, or completing code reviews. Triggers on "analyze for automation", "what ca...
Extract text and tables from PDF files, fill forms, merge documents.
Analyze datasets and generate insights from data.
Gmail-integrated CRM for managing pipelines, deals (boxes), contacts, and email threads
Reportei marketing report generation API via curl. Use this skill to manage clients, reports, templates, integrations and webhooks for automated marketing analytics.
Resend email API via curl. Use this skill to send transactional emails, manage contacts, domains, and API keys.
Open-source CRM API for managing people, companies, notes, tasks and custom objects
Bright Data Web Scraper API via curl. Use this skill for scraping social media (Twitter/X, Reddit, YouTube, Instagram, TikTok), account management, and usage monitoring.
Monday.com GraphQL API via curl. Use this skill to manage boards, items, and projects.
Dev.to API via curl. Use this skill to publish and manage articles on Dev.to.
Gmail API via curl. Use this skill to read, send, and manage emails, labels, drafts, and threads.
Lark/Feishu API integration for messaging, group management, contacts, and calendar. Use this skill to send messages, manage chats, query contacts, and sync calendar events with Lark/Feishu.
Upload images to Imgur for free hosting. Use this skill when you need to upload images and get public URLs for sharing or embedding in articles.
Google Sheets API via curl. Use this skill to read, write, and manage spreadsheet data programmatically.
Figma REST API for accessing design files, comments, components, and projects. Use this skill to read file contents, export images, manage comments, and integrate with Figma workspaces.
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.