# Anthropic open-sources 11 Claude plugins for knowledge work roles

> **Key takeaway:** Anthropic has released an open-source repository of 11 plugins that extend Claude with role-specific skills, slash commands, and tool connectors for jobs like sales, product management, data analysis, and bio-research.

**Published:** 2026-10-23T06:03:06.040214+00:00 
**Source:** GitHub Trending (daily) 
**Category:** foss-apps 
**Canonical URL:** https://fosswire.org/news/anthropic-open-sources-11-claude-plugins-for-knowledge-work-.html 

## Summary
Anthropic published knowledge-work-plugins, a GitHub repository containing 11 open-source plugins for its Claude Cowork and Claude Code products. Each plugin bundles domain-specific skills, slash commands, and MCP-based connectors to external tools like Slack, Jira, Snowflake, and PubMed. The plugins cover roles including productivity, sales, customer support, product management, marketing, legal, finance, data, enterprise search, bio-research, and plugin management itself. All configuration is file-based using markdown and JSON with no code or build steps required.

## Plugin structure and customization
- Each plugin uses a standard directory structure: `.claude-plugin/plugin.json` manifest, `.mcp.json` for tool connections, `commands/` for slash commands, and `skills/` for automatic domain knowledge
- Connectors use MCP (Model Context Protocol) servers to interface with external tools
- Components are file-based (markdown and JSON) with no code, infrastructure, or build steps
- Installation for Claude Code uses: `claude plugin marketplace add anthropics/knowledge-work-plugins` then `claude plugin install <plugin>@knowledge-work-plugins`
- Skills activate automatically when relevant; slash commands follow the pattern `/<role>:<action>` such as `/sales:call-prep` or `/data:write-query`

## Why it matters
The release provides a concrete, inspectable example of how enterprises can structure AI agent capabilities for specific roles without writing code, potentially accelerating adoption of MCP as a standard for AI-tool integration.