AI Agent Index
ByHeather MacAvelia·Independently reviewed·Last verified Jul 10, 2026

Anthropic conversational AI powered by Fable 5, Opus 4.8, Sonnet 5, Sonnet 4.6, and Haiku 4.5. Research, Code, Cowork, Design, Managed Agents, MCP Connectors, Office add-ins. Free; Pro $17/mo annual.

How we scored it

Autonomy

4/5

Integrations

5/5

Pricing clarity

5/5

Evidence

5/5

Setup

5/5

The facts

Claude is the conversational AI platform from Anthropic, available at claude.ai on web, iOS, Android, macOS, and Windows. The Claude model family currently available includes Fable 5 (Mythos-class frontier model restored July 1, 2026 after export control suspension, $10/$50 per million tokens input/output), Opus 4.8 (complex reasoning and agentic coding, released May 28, 2026), Sonnet 5 (near-Opus performance at lower cost, launched June 30, 2026, $2/$10 per million tokens introductory through August 31), Sonnet 4.6 (balanced intelligence and speed), and Haiku 4.5 (fastest and most cost-efficient). Fable 5 was originally launched June 9, 2026, suspended June 12 under a US government export control directive, and restored July 1 after the controls were lifted. Fable 5 is available on the Claude Platform, Claude.ai, Claude Code, and Claude Cowork. For Pro, Max, Team, and select Enterprise plans, Fable 5 is included for up to 50% of weekly usage limits through July 7, after which standard usage-based pricing applies. Fable 5 includes safety classifiers that block certain cybersecurity and biology requests. Blocked requests return an error rather than a response and are not charged. Inputs and outputs from Fable 5 are retained for 30 days to support safety work. Managed Agents includes built-in fallback. API users can opt in to automatic fallback to Opus 4.8 for blocked requests. Beyond conversational chat, Claude includes agentic features. Research runs multi-step autonomous web investigations. Claude Managed Agents (public beta) provides a harness for long-running agentic work. Projects offer persistent context and document workspaces. Memory enables cross-session recall. Skills are organization-deployable custom capabilities. MCP Connectors integrate any external data source or tool via Model Context Protocol. Artifacts provide live code execution and file creation. Parallel sub-products ship within the same Pro plan. Claude Code provides terminal-based agentic coding. Claude Cowork is a knowledge-work desktop app. Claude Design creates visual outputs including prototypes, slides, and designs. Claude for Excel, PowerPoint, and Word covers Office integrations. Claude for Chrome is a browsing agent. Claude Tag lets Team and Enterprise users tag Claude directly in Slack conversations to delegate tasks. Claude Science launched June 30, 2026 as a customizable AI workbench for scientists. NASA used Claude for the first AI-assisted drive on another planet. The platform serves knowledge workers, researchers, developers, and enterprise teams globally. Claude pioneered the Model Context Protocol (MCP) as a first-class connector system, enabling integration with any external data source or tool through remote MCP without custom development. Native connectors include Gmail, Google Calendar, Google Drive, and Slack. Office add-ins cover Excel, PowerPoint, and Word. Claude deploys across the Anthropic API (direct), Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry. Claude for Chrome provides browser automation. Claude Code integrates directly with local codebases and terminal environments. The MCP-native architecture creates a genuine integration moat: power users and enterprise teams can connect Claude to proprietary data sources, internal tools, and custom workflows without waiting for each connector to be built. Pricing is fully public with a permanent free tier. Free ($0) includes basic chat, web search, code and data analysis, MCP connectors, and extended thinking for complex work. Pro ($17/month annual or $20/month monthly) adds Claude Code, Cowork, Design, Excel/PowerPoint integrations, unlimited Projects, Research mode, Memory, and higher usage limits. Max (from $100/month) provides 5x or 20x Pro usage with priority access and early access to advanced features. Team Standard ($20/seat annual) and Team Premium ($100/seat annual) for 5 to 150 users add SSO, central admin, enterprise search, and Claude Tag in Slack. Enterprise ($20/seat plus usage at API rates) adds SCIM, audit logs, Compliance API, HIPAA-ready deployment, custom data retention, network access controls, and IP allowlisting. API pricing is separate and per-token. Claude does not offer image or video generation: users needing visual creation must pair Claude with Midjourney, DALL-E, Sora, or similar tools. The third-party app ecosystem is smaller than the ChatGPT Custom GPT marketplace. Rate limits at Pro ($17/month) are tighter than competing tiers at similar prices. Power users routinely hit limits on Opus and need Max ($100+/month) for sustained heavy daily use. ChatGPT Plus ($20/month) offers image generation and a larger plugin ecosystem. Gemini ($19.99/month) integrates natively with Google Workspace. Microsoft 365 Copilot integrates with the Microsoft 365 suite. Current state Q3 2026: Claude Fable 5 and Mythos 5 export controls were lifted June 30, 2026. Fable 5 access restored globally July 1 across Claude Platform, Claude.ai, Claude Code, and Cowork. Fable 5 includes safety classifiers, automatic model fallback to Opus 4.8 for blocked requests, and a 30-day data retention requirement. Claude Sonnet 5 launched June 30 as the most agentic Sonnet model, with near-Opus 4.8 performance at $2/$10 per million tokens (introductory through August 31). Claude Science launched as a customizable AI workbench for scientists. Claude Managed Agents remain in public beta with Fable 5 support and built-in fallback. Anthropic proposed an industry-wide jailbreak severity scoring framework with Amazon, Microsoft, Google, and other Glasswing partners. G2 shows 371 reviews at 4.6/5 (claimed by Anthropic). SOC 2 Type II, ISO 27001, ISO 42001, HIPAA, and FedRAMP are confirmed on trust.anthropic.com.

Pricing

freemium · $17/mo annual

View pricing ↗

Segment

b2b

Setup

easy

Verified

Jul 10, 2026

Transparency

Public

Contract

Monthly or Annual

Data training

Opt-out

Autonomy

Human Optional

Capabilities

content-creationdeep-researchweb-searchdata-analysiscode-generationcitationsautonomousworkflow-builder

Pros & Limitations

Editorial assessment

Pros

  • Comprehensive sub-product suite (Claude Code, Cowork, Design, Managed Agents, Excel/PowerPoint/Word, Chrome, Slack) is included in a single Pro subscription at $17/month annual. Meaningfully better unit economics than purchasing separate subscriptions for chat, coding, research, design, and office productivity AI tools.
  • MCP-native architecture: Claude was the first platform to make Model Context Protocol a first-class connector system, letting users integrate any external data source or tool through remote MCP without custom development. This creates a real moat for power users and enterprise teams building extensible AI workflows.
  • Opus 4.8 remains a top-tier reasoning and agentic coding model with Research mode running multi-step autonomous web investigations. Claude Managed Agents (public beta) provides a harness for long-running agentic work across complex tasks.

Limitations

  • Fable 5 ships with safety classifiers that block certain cybersecurity and biology requests, and inputs and outputs are retained for 30 days to support safety work. Teams in sensitive domains may see blocked requests routed to the Opus 4.8 fallback rather than served by the top model.
  • No image or video generation native to the platform: users needing visual creation must pair Claude with Midjourney, DALL-E, Sora, or similar tools, breaking workflow continuity for teams that need both text and visual outputs in a single interface.
  • Rate limits at Pro ($17/month) are tighter than competing tiers at similar prices. Power users routinely hit limits on Opus and need to upgrade to Max ($100+/month) for sustained heavy daily use, creating a significant cost gap between casual and power user tiers.

Technical Details

Deployment
webmobiledesktop
Model architectureAnthropic Claude (Fable 5, Opus 4.8, Sonnet 5, Sonnet 4.6, Haiku 4.5)
Avg setup timeUnder 1 minute (sign up at claude.ai)
Autonomous rateResearch mode runs multi-step autonomous web investigations with planning, execution, and synthesis. Claude Code executes agentic coding workflows including file management, testing, and deployment from the terminal. Skills and MCP Connectors enable autonomous tool use within sessions. Cowork handles knowledge-work tasks on the desktop. Core conversational mode is human-directed with the user guiding each exchange.
MCP compatibleYes
Integrations
GmailGoogle CalendarGoogle DriveSlackMicrosoft ExcelMicrosoft PowerPointMicrosoft WordChrome ExtensionAmazon BedrockGoogle Cloud Vertex AIMicrosoft FoundryMCP Server
Security
SOC 2 Type IIISO 27001ISO 42001HIPAAFedRAMP

Similar agents

Rating

4.7/ 5

Editorial score

How we score this →

Recognition

Rated 4.7 ★Category Leader · Q3 2026Transparent PricingListed 2026
Embed these on your site →

Industries

B2BEnterpriseSMBSaaSAgenciesStartupsFinanceLegalHealthcare

Leave a review

Never displayed publicly.

Agent Stacks

See workflow stacks that feature Claude.

Compare

Related Content

Is this your tool?

Claim this listing free to get your embeddable badges, a review collection link, and the Verified checkmark.

Claim listing →

Free · Every Two Weeks

AI Agent Price & Rating Tracker

Price changes, new agent launches, acquisitions, and rating updates across 344+ AI agents. Verified against live vendor data, not vendor marketing.

No spam. Unsubscribe anytime. We never share your email.