Skip to main content

AI Tools & Platforms (2026)

info

This content was developed with AI assistance and is regularly reviewed for accuracy.

Snapshot: October 2026. Model names, prices, and product features change monthly; What's New in AI is the maintained chronology.

In short: The leading general-purpose AI tools in 2026 are Claude (claude.ai), ChatGPT (chatgpt.com), and Gemini (gemini.google.com). All three run agents, use tools (web, code, files) by default, remember across chats, and advertise 1M-token context at the API level on their current frontier families. The month-by-month release history behind those names lives in What's New in AI (2026); this page tells you what to use now. Pick one assistant as your daily driver, then add the specialist agents, AI IDEs, and creative tools below as you need them. Find your task in the table to jump to the right tool.

If you want to...Start withWhy
Ask, write, and research every dayClaude, ChatGPT, or GeminiOne general assistant covers most tasks; the comparison helps you choose
Write or fix codeClaude Code or CursorA coding agent in your terminal, or an AI-first code editor
Build a web app without codingv0.app or Bolt.newDescribe the app and get working code you can deploy
Hand off tasks in your browserClaude in ChromeFills forms and pulls data from web pages for you (paid Claude plans)
Automate work across your appsZapier AgentsDescribe a workflow in plain words and it builds and runs it, with a free tier
Research with citationsPerplexity or ElicitSourced answers for current events, or academic papers
Make imagesMidjourney or Canva AIArtistic images, or ready-to-post marketing graphics
Make videoRunwayAI video creation and editing, with a free tier
Transcribe meetingsOtter.aiLive transcripts and summaries of calls and lectures
Create voiceoversElevenLabsNatural voices in many languages
Pay nothingFree AI Tools DirectoryWhat each free tier includes and where its limits are
Build a practical stack first

Before you choose from this full list, read Modern AI Stack (2026) to set up a lean toolchain you can actually maintain.

Just here to pick a tool?

This page is a full reference, so it's fine to jump straight to the section you need. If you came to make a decision, these pages are quicker:

What's Changed Since Early 2025​

The Big Shifts:

  • 🤖 Chat → Agents - AI now completes tasks, not just answers questions
  • 🔧 Tools are standard - Web browsing, code execution, file access built-in
  • 📏 Context is huge - 1M-token windows are standard on the current frontier families (roughly 555,000 words on Anthropic's current tokenizer)
  • 🧩 Skills & MCP - Extend any assistant with reusable skills (Agent Skills open standard) and MCP connectors
  • 🎯 Multi-modal default - Text, images, audio, video all in one interface

What to expect:

  • Less copying/pasting outputs
  • More delegating complete workflows
  • AI that uses tools autonomously
  • Customization through skills and agents

Foundational AI Platforms​

Claude by Anthropic ⭐ Top Choice​

  • Link: claude.ai
  • Current lineup: Claude Fable 5.1 (frontier; $10/$50 per million input/output tokens), Opus 5.5 ($4/$20; Anthropic's recommended starting point for most work), Sonnet 5.5 ($2/$10), and Haiku 5.5 ($0.10/$0.50 for prompts up to 100K tokens). All four have a 1M-token context window at standard pricing and always-on adaptive thinking controlled by an effort setting rather than a "thinking mode" toggle. Release dates are in What's New in AI.
  • Pricing: Free (Sonnet and Haiku, with web search, memory, connectors, voice, file creation, and up to 5 projects), Pro ($20/month, or $17/month billed annually), Max ($100 or $200/month, with matching monthly API credits), Team ($20/seat/month annual, $25 monthly; premium seats with Fable $100), Enterprise (custom)
  • Best for: Agent work across your files and apps, coding, analysis, research, long documents
  • Agent features:
    • Agent mode - The former Claude Cowork, now simply part of Claude: give it a goal and it works across your files, connectors and a built-in browser, showing each step and asking before it deletes anything
    • Claude Code - Coding agent in the terminal, IDEs, desktop app, web and mobile
    • Scheduled tasks and task handoff - Hand a task to Claude from your phone or schedule it to run unattended (Pro and up)
    • Skills and MCP connectors - Reusable SKILL.md packages that follow the open Agent Skills standard, plus tool and data connections through the Model Context Protocol
    • Projects and memory - Persistent context across conversations, with memory on the free tier
  • Context: 1M tokens on all four current models (roughly 555,000 words on the current tokenizer)
  • Why choose: Best for complex reasoning, agentic workflows, and developer tasks

ChatGPT by OpenAI​

  • Link: chatgpt.com
  • Current lineup: GPT-6 Astra (flagship, rolled out in phases from September 2026 with its most advanced capabilities restricted to vetted cybersecurity partners), GPT-6.1 Sol and GPT-6 Sol (the paid-tier defaults in regular Chat since October 7, 2026), and GPT-6 Luna (the Free and Go default). Reasoning is built into every tier; there is no separate o-series. OpenAI's API prices, exact context windows, and the "GPT-6 Pro" label some reports give Astra in the Chat interface could not be confirmed against OpenAI's own pages for this snapshot, so treat third-party figures as reported. Release history: What's New in AI.
  • Pricing: Free (shows ads since February 2026), Go ($8/month in the US, reported), Plus ($20/month), Pro ($100 and $200/month, plus a reported $500 tier), Business (formerly Team; $20/user/month annual, $25/user/month billed monthly), Enterprise (custom)
  • Best for: General use, creative tasks with images and voice, quick answers, the broadest all-in-one feature set
  • Agent features:
    • ChatGPT Work - Agent mode (July 2026) that completes multi-step office tasks across connected apps on paid plans except Free and Go
    • Dots - Always-on agents for ongoing tasks (September 2026), reported for Pro tiers and Business Premium
    • Codex - Agentic coding in the cloud, the CLI and IDEs
    • Workspace Agents - Shared assistants for Business and Enterprise workspaces; they replaced Custom GPT creation for individuals in August 2026
    • Voice, image generation and Intelligent UI - Spoken conversation, the GPT Image family, and answers with interactive charts, forms and buttons
  • Context: 1M tokens advertised at the API level for the GPT-6 family (reported; a given product's effective limit can be lower)
  • Why choose: Largest user base, most familiar interface, images and voice in one app

Gemini by Google​

  • Link: gemini.google.com
  • Current lineup: Gemini 4 Argon (flagship; in the Gemini app for Google AI Pro and Ultra subscribers and through the paid API since October 13, 2026), Gemini 3.1 Pro, Gemini 3.8 Flash (the fast workhorse, which Google says beats 3.1 Pro on some agentic tasks), and Gemini 3.5 Flash-Lite (the free-tier model since October 9, 2026, reported). Deep Think is reserved for higher paid tiers. Gemini 4 Carbon (announced October 28, 2026) is limited to vetted cyber defenders for now and isn't available to subscribers. Release history: What's New in AI.
  • Pricing: Free (Flash-Lite only, reported), Google AI Plus (about $5/month, reported; adds Flash), Google AI Pro ($20/month, formerly Gemini Advanced; adds Gemini 3.1 Pro and Gemini 4 Argon, plus 2TB of storage), Google AI Ultra ($100 and $200/month, reported). The standalone Gemini Business/Enterprise add-ons were discontinued in 2025; Gemini is bundled into Google Workspace plans (Business Standard ~$14/user/month, Business Plus ~$22/user/month, annual billing)
  • Best for: Google Workspace users, multimodal tasks, real-time information
  • Agent features:
    • Agent Mode - Early-access agent features on paid tiers
    • Gemini agent for work - One agent across Gmail, Drive, Docs and Chat, in enterprise private preview (announced October 2026, reported)
    • Gems and Opal - Custom instructions and personas; Opal turns a Gem into a shareable no-code mini-app
    • Google Workspace - Native Gmail, Docs, Sheets integration, and Deep Research that can read your Gmail and Drive with permission
    • Multimodal native - Text, images, video, audio in one prompt
  • Context: 1M+ tokens advertised on 3.1 Pro and 3.8 Flash; Google quotes a 1M-token output limit for Argon (its input window was not confirmed)
  • Why choose: Best Google integration, excellent for research, strong multimodal capabilities

AI Agents & Autonomous Tools​

The new frontier: AI that completes tasks autonomously, not just chat.

Claude Code by Anthropic ⭐ Developer Favorite​

  • Link: code.claude.com
  • Description: Agentic coding across your whole codebase, in the terminal, VS Code and JetBrains, a desktop app, the web, and the mobile app
  • Pricing: Included with Claude Pro, Max, Team and Enterprise (usage counts against plan limits)
  • Best for: Software development, debugging, large refactors, project scaffolding
  • What it does:
    • Reads and writes files, runs commands and tests, uses git
    • Follows project rules in CLAUDE.md (and reads AGENTS.md in its place)
    • Delegates to subagents, agent teams, and scripted workflows
    • Keeps working on a schedule or toward a goal with /loop, /goal, and cloud Routines
    • Reviews pull requests with Code Review (generally available October 15, 2026 for Team and Enterprise)
  • Adoption: Stack Overflow's 2026 survey found 66% of coding-agent users use Claude Code, ahead of GitHub Copilot (59%), Codex (30%) and Cursor (25.5%)
  • Example: "Build a REST API with user authentication" → Claude Code plans, implements, tests and opens the pull request
  • Learn more: Claude Code: Your AI Coding Agent
  • Link: cursor.com
  • Description: AI-first code editor (VS Code fork) with agent mode
  • Pricing: Free tier, Pro ($20/month — now funds a monthly usage-credit pool rather than unlimited fast requests), Teams (formerly "Business," ~$40/user/month — verify current terms on Cursor's pricing page)
  • Best for: Day-to-day coding, refactoring, learning new codebases
  • Agent features:
    • Agent mode - Multi-file edits and reasoning
    • Codebase understanding - Knows your entire project
    • Composer - Natural language to implementation
    • Terminal integration - Run and debug automatically
  • Why it won: Familiar VS Code interface + powerful AI integration

GitHub Copilot Coding Agent​

  • Link: github.com/features/copilot
  • Description: Task-based development agent inside the GitHub Copilot product line (issue → plan → code → PR).
  • Pricing: Included with Copilot Pro+, Business, and Enterprise plans
  • Best for: Feature development, bug fixes, team workflows tied to GitHub Issues
  • Workflow:
    1. Assign a GitHub issue to Copilot
    2. Agent creates implementation plan
    3. Generates code across files
    4. Opens a pull request
    5. Responds to review feedback
  • Status: Generally available; check the GitHub Copilot pricing page for the latest plan-by-plan availability.

Devin by Cognition​

  • Link: devin.ai
  • Description: Autonomous AI software engineer
  • Pricing: Core $20/month (pay-as-you-go usage billing), Team $500/month, Enterprise (custom)
  • Best for: Complex projects, junior developer augmentation
  • Capabilities:
    • Builds full applications from scratch
    • Debugs in sandboxed environments
    • Learns new technologies autonomously
    • Collaborates via chat interface
  • Note: With Devin 2.0, Devin dropped its enterprise-only positioning — the Core plan is now accessible to individual developers, not just teams

Replit Agent​

  • Link: replit.com/agent
  • Description: Natural language to full working application
  • Pricing: Free tier, Replit Core ($25/month) for higher compute and AI usage
  • Best for: Rapid prototyping, learning, MVPs
  • Example: "Build a todo app with Firebase auth" → Complete deployed app in minutes

Claude in Chrome (Browser Agent)​

  • Link: Chrome Web Store
  • Description: Browser automation and web research agent
  • Pricing: Included with all paid Claude plans; on Max and Team it runs as a full agent session, rolling out to Pro
  • Best for: Research, data extraction, web automation, form filling (Anthropic advises starting with trusted sites and keeping it away from financial, legal and medical sites)
  • Can do:
    • Browse and research autonomously
    • Fill forms and interact with sites
    • Extract structured data
    • Monitor websites for changes
    • Complete multi-step web tasks
  • Example: "Research our 5 competitors and create comparison spreadsheet"

Perplexity Pro​

  • Link: perplexity.ai
  • What's new: The Comet browser on Windows, macOS, iOS and Android, a "Computer" agent on Max that drives it, Perplexity Pages, file analysis
  • Pricing: Free tier, Pro ($20/month), Max ($200/month, reported), Enterprise (custom)
  • Best for: Deep research, academic work, current events
  • Agent features: Can research multiple sources, synthesize findings, create shareable pages
  • Why choose: Best for research with citations, beats Google for complex questions

No-Code Agent Builders​

Build custom AI agents without coding.

Relevance AI ⭐ AI Employees​

  • Link: relevanceai.com
  • Description: Build autonomous "AI employees" for specific jobs
  • Pricing: Starter ($99/month), Pro ($249/month), Enterprise (custom)
  • Examples: Research agent, data enrichment agent, customer support agent
  • Use case: Replace repetitive human work with trained AI agents

Lindy​

  • Link: lindy.ai
  • Description: Personal AI assistant that takes actions across your tools
  • Pricing: Reported from about $30/month per user, with voice agents on Business and Enterprise tiers; check lindy.ai for current terms
  • Best for: Personal productivity, meeting prep, email management
  • Can do: Schedule meetings, draft emails, research topics, track tasks

Respell​

  • Link: respell.ai
  • Description: No-code AI agent and workflow builder
  • Pricing: Free tier, Pro ($29/month), Team plans available
  • Best for: Business process automation, custom workflows
  • Features: Visual workflow builder, pre-built templates, tool integrations

Zapier Agents​

  • Link: zapier.com/agents
  • Description: Natural language automation - describe what you want, AI builds and runs the workflow
  • Pricing: Free tier (reported 400 agent activities per month); paid Agents plans from about $33/month billed annually (reported, October 2026)
  • Evolution: Zapier evolved from trigger-action to natural language agents, and its MCP server lets Claude and other assistants reach the same app catalog
  • Example: "When I get tagged in Linear, research the issue and post summary to Slack"

AI Writing & Content Creation​

Specialized tools for content creation, copywriting, and editorial assistance.

note

By 2026, general-purpose models (Claude, ChatGPT, Gemini) have largely closed the gap with the dedicated copywriting platforms that defined 2022–2023. Tools like Jasper and Copy.ai still operate, but most teams now get comparable or better output from a general-purpose chat model with a good prompt or custom instructions. Consider the specialized tools below mainly if you need their workflow features (templates, brand-voice training, team collaboration) rather than raw copy quality.

Writesonic​

  • Link: writesonic.com
  • Description: AI writing assistant with real-time data and SEO optimization capabilities.
  • Pricing: Free tier, Pro ($16/month), Enterprise ($400/month)
  • Best for: SEO content, article writing, marketing copy
  • Key features: Real-time data, SEO optimization, bulk content generation, AI art

Grammarly​

  • Link: grammarly.com
  • Description: AI-powered writing assistant for grammar, tone, and style improvement. Now one product within the Superhuman suite (Grammarly, Superhuman Docs (formerly Coda), and Superhuman Mail), following its parent company's 2025 rebrand from Grammarly to Superhuman.
  • Pricing: Free tier, Premium ($12/month), Business ($15/user/month)
  • Best for: Writing improvement, editing, professional communication
  • Key features: Grammar checking, tone detection, plagiarism detection, browser extension, integrated Superhuman Go AI assistant

Notion AI​

  • Link: notion.so/product/ai
  • Description: AI writing assistant integrated into Notion's productivity platform.
  • Pricing: Included with the Business plan (~$20/user/month, annual billing); Notion discontinued the standalone $10/user AI add-on in 2025
  • Best for: Notion users, document creation, note enhancement
  • Key features: Native Notion integration, document analysis, content generation

AI Visual Creation​

Tools for generating, editing, and enhancing images, videos, and visual content.

Midjourney​

  • Link: midjourney.com
  • Description: Premier AI art generator known for high-quality, artistic image creation.
  • Pricing: Basic ($10/month), Standard ($30/month), Pro ($60/month)
  • Best for: Artistic images, concept art, creative projects, professional illustration
  • Key features: Discord interface, style consistency, high-resolution outputs, community gallery

GPT Image 2.5 by OpenAI​

  • Link: openai.com
  • Description: OpenAI's native image generation in ChatGPT and via the OpenAI API. The GPT Image family replaced DALL-E in 2026 and adds built-in reasoning, precise single-turn editing, and accurate text rendering; version history is in What's New in AI.
  • Pricing: Included with ChatGPT Plus/Business; usage-based API pricing
  • Best for: Precise prompt following, realistic images, ChatGPT integration, multi-turn image editing
  • Key features: Natural language prompts, in-chat editing, accurate text rendering, ChatGPT integration

Gemini Image ("Nano Banana" family) by Google​

  • Link: gemini.google.com
  • Description: Google's image generation models built into Gemini for text-to-image and image editing tasks, spanning Flash Lite, Flash, and Pro tiers. Nano Banana 2 (February 2026) made pro-quality generation fast and cheap, and point releases have followed (reported); the free tier includes rate-limited generation.
  • Pricing: Included with Gemini (free tier with limits, full access via Google AI Pro)
  • Best for: Multimodal editing, Workspace-integrated workflows
  • Key features: Native multimodal input, image editing, integration with Google Workspace

Stable Diffusion (Various Platforms)​

  • Link: stability.ai
  • Description: Open-source image generation model available through multiple platforms.
  • Pricing: Varies by platform (many free options available)
  • Best for: Customization, local deployment, cost-effective generation
  • Key features: Open source, highly customizable, multiple hosting options

Canva AI​

  • Link: canva.com
  • Description: AI-powered design features integrated into Canva's design platform.
  • Pricing: Free tier, Pro ($13/month), Teams ($20/user/month)
  • Best for: Marketing materials, presentations, social media graphics
  • Key features: Magic Design, background removal, text-to-image, template suggestions
Video in 2026

OpenAI shut down Sora (announced March 24, 2026; the app went offline in April and the API in September, reported). The video generators people reach for now are Google's Veo 3.1 inside Gemini and Flow, Kling 3.0, Seedance 2.0, and Runway. Dates and details are in What's New in AI.

Runway​

  • Link: runwayml.com
  • Description: AI video generation and editing platform with creative tools.
  • Pricing: Free tier, Standard ($12/month), Pro ($28/month), Unlimited ($76/month)
  • Best for: Video creation, AI filmmaking, creative experiments
  • Key features: Text-to-video, image-to-video, video editing, motion brush

Luma App (formerly Dream Machine)​

  • Link: lumalabs.ai
  • Description: AI video generation platform focused on realistic and cinematic outputs, built on Luma's Ray model family (Ray 3.2 as of mid-2026, with native 1080p and frame-level keyframe control).
  • Pricing: Free tier, Pro plans available
  • Best for: Realistic video generation, cinematic content
  • Key features: Text-to-video, image-to-video, camera control, high-quality outputs

AI Development Tools​

v0.app by Vercel ⭐ Frontend Agent​

  • Link: v0.app (renamed from v0.dev in January 2026)
  • Description: Natural language to full-stack apps, from React/Next.js UI components to backend and database
  • Pricing: Free tier, Pro ($20/month)
  • Best for: Frontend development, UI components, rapid prototyping
  • What it does: Describe a UI, get production-ready React code
  • Why it's big: Fastest way to go from idea to working frontend

Bolt.new by StackBlitz​

  • Link: bolt.new
  • Description: Full-stack applications from prompts, running in browser
  • Pricing: Free tier, Pro plans available
  • Best for: Full-stack MVPs, demos, learning
  • Magic: Entire apps with backend, database, UI generated and running
  • Example: "Build a link shortener with analytics" → Working app in minutes

GitHub Copilot​

  • Link: github.com/features/copilot
  • What's new: Multi-file editing, the Copilot coding agent, deeper IDE integration
  • Pricing: Pro ($10/month), Pro+ ($39/month), Max ($100/month) for individuals; Business ($19/user/month), Enterprise ($39/user/month) for teams — as of June 1, 2026, all tiers fund a monthly GitHub AI Credit allotment under usage-based billing rather than flat unlimited-ish usage; check GitHub's pricing page for current credit amounts
  • Best for: Code completion, GitHub-native workflows
  • Still relevant: Yes, but Cursor has taken market share for pure editing

Devin Desktop (formerly Windsurf, formerly Codeium)​

  • Link: devin.ai (Cognition's site; the old windsurf.com address no longer applies)
  • Description: Agentic IDE and code assistant with a generous free tier (Codeium's autocomplete extensions still work under the current brand)
  • Pricing: Free for individuals, paid plans add team collaboration and the Devin Local agent
  • Best for: Developers who want a free Copilot alternative or a full AI-first IDE
  • Why choose: Free autocomplete for solo devs, 70+ languages, and a path into the agent IDE when you need more
  • Note: Cognition (also the maker of Devin, profiled above) acquired Windsurf in July 2025, following a restructuring in which Google licensed key technology and hired much of the founding team, and rebranded it Devin Desktop in June 2026 (reported).

Development Frameworks & Tools​

LangChain → LangGraph​

  • Link: langchain.com
  • Evolution: LangChain for simple chains → LangGraph for complex agents
  • Best for: Building production AI agents with explicit control flow when you want to stay vendor-neutral
  • 2026 status: Widely used, but no longer the default. First-party options (the Claude Agent SDK, Microsoft Agent Framework) and the vendors' built-in subagents, workflows and MCP support cover many cases without a framework; see Patterns and Multi-Agent Systems

CrewAI​

  • Link: crewai.com
  • Description: Framework for orchestrating multiple AI agents
  • Best for: Multi-agent systems where specialized agents collaborate
  • Example: Research crew (search agent + analyst + writer)

Claude Agent SDK (Anthropic)​

  • Link: github.com/anthropics/claude-agent-sdk
  • Description: Official framework for building Claude-powered agents
  • Best for: Production agents with tool use and MCP integration
  • Status: Standard SDK for serious Claude-based agent development
  • Link: modal.com, e2b.dev
  • Description: Run AI agents in secure, isolated environments
  • Why you need it: Agents executing code need sandboxing
  • Use case: Safe code execution, testing, production deployments

AI Business & Productivity​

AI-powered tools for business operations, productivity, and workflow automation.

Monday.com AI​

  • Link: monday.com
  • Description: Project management platform with integrated AI for task automation and insights.
  • Pricing: Basic ($8/user/month), Standard ($10/user/month), Pro ($16/user/month)
  • Best for: Project management, team collaboration, workflow automation
  • Key features: AI task suggestions, automated updates, intelligent scheduling, data insights

Zapier AI​

  • Link: zapier.com
  • Description: Automation platform with AI-powered workflow creation and optimization.
  • Pricing: Free tier, Professional (from $19.99/month with annual billing, $29.99/month billed monthly), Team ($103.50/month), Enterprise (custom)
  • Best for: Workflow automation, app integration, repetitive task elimination
  • Key features: AI workflow suggestions, natural language automation, 5000+ app integrations

Clickup AI​

  • Link: clickup.com
  • Description: All-in-one productivity platform with AI writing and task management features.
  • Pricing: Free tier, Unlimited ($7/user/month), Business ($12/user/month)
  • Best for: Project management, document creation, team productivity
  • Key features: AI writing assistant, task automation, project templates, time tracking

Superhuman AI​

  • Link: superhuman.com
  • Description: AI-powered email client focused on speed and efficiency, and the product that gave the Superhuman brand its name after Grammarly's parent company rebrand (see the Grammarly entry above).
  • Pricing: $30/user/month
  • Best for: Email power users, sales teams, executive assistants
  • Key features: AI email writing, smart scheduling, lightning-fast interface, email insights

Otter.ai​

  • Link: otter.ai
  • Description: AI meeting transcription and note-taking service with real-time collaboration.
  • Pricing: Free tier, Pro ($17/month), Business ($20/user/month)
  • Best for: Meeting transcription, interview recording, lecture notes
  • Key features: Real-time transcription, meeting summaries, speaker identification, integrations

Krisp​

  • Link: krisp.ai
  • Description: AI-powered noise cancellation and meeting enhancement tool.
  • Pricing: Free tier, Pro ($8/month), Business ($5/user/month)
  • Best for: Remote work, call quality improvement, meeting transcription
  • Key features: Noise cancellation, voice clarity, meeting transcripts, accent localization

AI Research & Analysis​

Tools for data analysis, research, and information synthesis.

Elicit​

  • Link: elicit.org
  • Description: AI research assistant for academic literature review and analysis.
  • Pricing: Free tier, Plus ($10/month), Pro ($42/month)
  • Best for: Academic research, literature reviews, scientific analysis
  • Key features: Paper search, systematic reviews, data extraction, research synthesis

Consensus​

  • Link: consensus.app
  • Description: AI-powered search engine for scientific research with evidence-based answers.
  • Pricing: Free tier, Premium ($8.99/month)
  • Best for: Scientific research, evidence-based conclusions, academic writing
  • Key features: Scientific paper search, evidence synthesis, citation tracking, bias detection

Research Rabbit​

  • Link: researchrabbit.ai
  • Description: AI-powered research discovery platform for finding and organizing academic papers.
  • Pricing: Free
  • Best for: Academic research, paper discovery, research collaboration
  • Key features: Paper recommendations, visual research maps, collaboration tools, citation networks

Specialized AI Applications​

Niche AI tools for specific industries and use cases.

Descript​

  • Link: descript.com
  • Description: AI-powered audio and video editing with text-based editing interface.
  • Pricing: Free tier, Creator ($12/month), Pro ($24/month), Enterprise ($40/user/month)
  • Best for: Podcast editing, video content creation, voice cloning
  • Key features: Text-based editing, voice cloning, transcription, overdub technology

Beautiful.AI​

  • Link: beautiful.ai
  • Description: AI-powered presentation creation with intelligent design assistance.
  • Pricing: Free tier, Pro ($12/month), Team ($40/user/month)
  • Best for: Presentation design, pitch decks, business presentations
  • Key features: Smart slide templates, automatic formatting, design suggestions, team collaboration

Murf​

  • Link: murf.ai
  • Description: AI voice generation platform for creating professional voiceovers.
  • Pricing: Free tier, Basic ($19/month), Pro ($26/month), Enterprise (custom)
  • Best for: Voiceovers, e-learning content, marketing videos
  • Key features: 120+ voices, voice cloning, emotion control, multi-language support

ElevenLabs​

  • Link: elevenlabs.io
  • Description: Advanced AI voice synthesis with realistic and expressive speech generation.
  • Pricing: Free tier, Starter ($5/month), Creator ($22/month), Pro ($99/month)
  • Best for: High-quality voice synthesis, audiobook creation, multilingual content
  • Key features: Voice cloning, emotion and style control, 29+ languages, API access

Skills & Customization Ecosystem​

Agent Skills (open standard) ⭐​

  • Link: agentskills.io
  • Description: A skill is a folder with a SKILL.md file (only name and description are required) that packages instructions, reference files, and scripts an agent loads on demand
  • Status (October 2026): Released as an open standard in December 2025. By March 2026 OpenAI, Microsoft, JetBrains, Cursor, Google's Gemini CLI, Block's Goose and dozens of other agent products had shipped compatible implementations, and OpenAI maintains its own public skills repository (reported)
  • Why it matters: Write a skill once and reuse it across Claude, Codex, Copilot and any other agent that reads the format
  • Learn more: Skills, AGENTS.md, and CLAUDE.md

Custom GPTs → Workspace Agents (ChatGPT)​

  • Link: chatgpt.com
  • Status (October 2026): Individuals can no longer create Custom GPTs (a change made in August 2026); OpenAI replaced them with Workspace Agents for Business and Enterprise workspaces. There is no creator revenue-sharing programme to plan around
  • What to do instead: For personal reuse, use Projects (ChatGPT or Claude) or Gems (Gemini); for shared team assistants, use Workspace Agents or Claude Projects; for portability across vendors, write an Agent Skill
  • Examples: Brand-voice writer, policy Q&A, code reviewer, onboarding guide

Claude Skills (Anthropic)​

  • Link: Claude apps, Claude Code, or the API
  • Description: Claude's implementation of the Agent Skills standard; skills sync between claude.ai and Claude Code
  • Integration: Pairs with MCP connectors (skills carry the know-how, connectors carry the tools)
  • Best for: Repeatable workflows such as brand voice, report formats and review checklists; developer workflows; enterprise customization
  • Format: SKILL.md with instructions plus optional reference files and scripts

MCP (Model Context Protocol) 🔧​

  • Link: modelcontextprotocol.io
  • What it is: The open standard for connecting AI assistants and agents to tools and data; for end users it shows up as "connectors"
  • Governance: Donated to the Linux Foundation's Agentic AI Foundation in December 2025 (co-founded by Anthropic, OpenAI and Block). The current spec revision, 2026-07-28, moved to a stateless HTTP transport, deprecated the older SSE transport, hardened OAuth with Client ID Metadata Documents, and added a Tasks extension
  • Finding servers: Anthropic's original reference servers for GitHub, Slack, Postgres and Google Drive were retired. Prefer the vendor's own first-party server (GitHub, Slack and Google Workspace each publish one), a connector from a reviewed directory such as the one inside Claude, or a currently maintained entry in the MCP Registry, and verify every server before connecting
  • Learn more: MCP Explained: Connectors for Users, Protocol for Builders

Getting Started: 2026 Edition​

Complete Beginners​

Start here:

  1. Claude or ChatGPT - Pick one, learn the basics (they're both excellent)
  2. First agent task: "Research X topic and create a summary document"
  3. Learn the method: Start Here: You Already Use AI teaches decide → brief → supply context → check → protect
  4. Write a better request: Brief and Prompt Library

Developers (2026 Path)​

Essential stack:

  1. Code editor: Cursor, or VS Code or JetBrains with Copilot or the Claude Code extension
  2. Terminal agent: Claude Code (the most-used coding agent in Stack Overflow's 2026 survey) or Codex
  3. Rapid prototyping: v0.app (frontend) + Bolt.new (full-stack)
  4. Framework: Claude Agent SDK or LangGraph for custom agents
  5. Sandboxing: Modal or E2B for safe code execution

Progression:

  • Week 1: Use Cursor for daily coding
  • Week 2: Try Claude Code for a full feature
  • Week 3: Build something with v0.app or Bolt.new
  • Week 4: Create your first custom agent

Business Users (2026 Path)​

Shift to agents:

  1. Personal agent: The agent mode in the assistant you already pay for (Claude, ChatGPT Work, Gemini Agent Mode); see Agents in the Apps You Already Pay For. Lindy or Relevance AI if you want a dedicated product
  2. Workflow automation: Zapier Agents or Claude scheduled tasks (natural language → recurring automation)
  3. Research agent: Claude Research or Perplexity for competitive intelligence
  4. Team setup: Claude Team or ChatGPT Business for collaboration

Think in agents, not tools:

  • Instead of "AI helps me write emails"
  • Think "Email agent handles routine correspondence"

Content Creators (2026 Path)​

Agent workflow:

  1. Research agent: Claude in Chrome or Perplexity
  2. Content agent: Claude with custom writing skill
  3. Visual agent: Midjourney, GPT Image 2.5 (ChatGPT), or Nano Banana (Gemini)
  4. Distribution: Automation via Make or Zapier
  5. Analytics: Claude analyzing performance data

Example full workflow: "Create weekly newsletter" → Research agent finds content → Writing agent drafts → Design agent creates images → Distribution agent schedules → Analytics agent reports performance

Cost Management Tips​

Free Tier Strategy:

  • Most tools offer generous free tiers - start there
  • Rotate between free trials to test multiple tools
  • Use free alternatives like Devin Desktop (formerly Windsurf/Codeium) instead of paid options

Subscription Optimization:

  • Many tools offer annual discounts (typically 20-30% off)
  • Team plans often provide better per-user value
  • Cancel subscriptions you're not actively using - most allow easy re-activation

Usage Monitoring:

  • Track your actual usage vs. plan limits
  • Consider pay-as-you-go options for sporadic use
  • Set spending alerts where available

Safety Considerations​

warning

Important Safety Guidelines

  • Data Privacy: Review privacy policies before uploading sensitive information
  • Output Verification: Always fact-check AI-generated content
  • Bias Awareness: Be mindful that AI outputs may reflect training data biases
  • Professional Review: Have human experts review critical content

Best Practices:

  • Test tools with non-sensitive data first
  • Understand each tool's data retention policies
  • Use workspace/team features to control data access
  • Keep backups of important work created with AI tools
  • Stay informed about each platform's terms of service changes

Next steps​