claude-cowork capabilities
20 mapped capabilities, each graded and dated. This is the diagnosis — the migration guide is the cure.
Capabilities
Action item extraction
provisionalverified 4 days agoExtract, organize, and format action items from meeting transcripts and documents.
Agentic task execution
provisionalverified 4 days agoAutonomous multi-step workflow execution on files, documents, and data through the Claude desktop app.
Artifacts
provisionalverified 4 days agoCowork 'live artifacts' are persistent, interactive HTML pages (trackers, dashboards, comparison tools, morning briefs) that survive across sessions and can call the user's MCP connectors at open time to repaint with fresh data. They are distinct from claude.ai chat Artifacts, which are single-session and don't call MCPs live.
Audience and product shape
provisionalverified 4 days agoClaude Cowork is Anthropic's desktop agent for non-developer knowledge workers: it brings Claude Code's agentic, multi-step task execution into the Claude Desktop app so people in sales, finance, legal, marketing, ops, HR, and design can hand off complex work that touches local files, web data, and connected SaaS tools.
Connected folder
provisionalverified 4 days agoCowork operates on user-selected local folders that the user explicitly grants Claude read/write/edit access to. Claude can read every file inside, write new files, and modify existing ones, producing real .docx/.xlsx/.pptx/.pdf outputs in place rather than chat-only artifacts.
Desktop chat window
provisionalverified 4 days agoThe primary surface is the Claude Desktop app, which has a mode selector toggling between 'Chat' and 'Cowork' (Tasks). Cowork tasks render inline in the desktop window with a sidebar for tasks, scheduled jobs, live artifacts, and the Customize menu.
Local file system access
provisionalverified 4 days agoRead, write, and organize files on the user's local file system through the Claude desktop app.
MCP connectors
provisionalverified 4 days agoExternal SaaS access in Cowork is delivered through MCP (Model Context Protocol) connectors — remote MCP servers the user authorizes via OAuth that let Claude read and act in tools like Google Drive, Gmail, Slack, Asana, Notion, Supabase, Vercel, Airtable, and more.
Meeting notes synthesis
provisionalverified 4 days agoGenerate structured meeting summaries from uploaded transcripts, recordings, or raw notes.
Model choice
provisionalverified 4 days agoCowork uses Anthropic's frontier models with a shared picker (Opus / Sonnet / Haiku) — the same model lineup users see in regular Claude chat, with Opus reserved for harder reasoning and Sonnet as the default workhorse.
Multi-source analysis
provisionalverified 4 days agoAnalyze and cross-reference information across multiple documents, files, and data sources in a single task.
Onboarding role picker
provisionalverified 4 days agoFirst-run flow for Cowork that asks the user what role they have (sales, finance, marketing, ops, etc.) and uses that to recommend a starter set of plugins, connectors, and Skills tailored to that function — the bridge between 'fresh install' and 'useful agent.'
Plans and pricing
provisionalverified 4 days agoCowork is gated behind paid Claude plans only — Pro ($20/mo), Max 5x (~$100/mo), Max 20x (~$200/mo), Team, and Enterprise. Free Claude users do not get Cowork. Originally Max-only at January 2026 launch, Pro access was added shortly after with lower usage limits than Max.
Plugins
provisionalverified 4 days agoPlugins are role-shaped bundles that package Skills, MCP connectors, and sub-agents together so a user gets a working setup from the first conversation instead of wiring pieces individually. Anthropic ships a library covering sales, finance, legal, marketing, HR, engineering, design, operations, data analysis, and more.
Projects
provisionalverified 4 days agoProjects in Cowork are persistent, self-contained workspaces that group related tasks, files, links, instructions, and memory together — the unit of organization for recurring or long-running work like 'Q3 board prep' or 'monthly close.'
Sandboxed Linux VM
provisionalverified 4 days agoEvery Cowork session runs shell commands and code inside an isolated per-session Linux VM (Ubuntu-based) with filesystem and network sandboxing, separate from the host OS. The user's connected folders are mounted in, and Claude executes bash there rather than on the host directly.
Scheduled tasks
provisionalverified 4 days agoScheduled tasks let a user describe a recurring job once — e.g. 'every Monday at 9am, build a weekly metrics report from these connectors' — and have Cowork run it automatically on that cadence without re-prompting.
Skills
provisionalverified 4 days agoSkills are Markdown-based instruction packs (SKILL.md plus optional assets) that give Claude a repeatable workflow or specialized knowledge for a task, loaded automatically when the model decides it's relevant. They appear in Cowork via plugins or as individually installed Anthropic Skills.
Slash commands
provisionalverified 4 days agoSlash commands are user-triggered shortcuts in the Cowork composer that invoke a structured workflow or built-in action. They are the manual counterpart to Skills — the user types / and picks from installed commands to kick off a specific, named workflow.
Sub-agents
provisionalverified 4 days agoCowork can split work across parallel sub-agents via a Task tool, letting the main agent fan out independent subtasks (research, file processing, parallel writes) and aggregate results, turning serial work like 'analyze 10 files' into concurrent work.