[GitHub Trending] anthropics/knowledge-work-plugins

6.5 relevance

Anthropic's knowledge work plugins for Claude are directly relevant to AI agent workflows.

Open Source github.com

Answer the board on AI readiness

Arrive with evidence, not opinion. See how your foundation benchmarks against your peers, get a clear read, and leave with a ranked place to start.

Why It Matters

For platform engineers and agent orchestrators, these open-source plugins provide a reference architecture for building reusable, domain-tuned sub-agents that integrate structured MCP connectors with implicit skills, reducing the gap between generic LLM assistance and production-ready, company-specific agent behavior.

Key Takeaways

Adopt the plugin pattern of pairing `.mcp.json` tool definitions with separate `skills/` and `commands/` directories to decouple domain expertise from orchestration in your own multi-agent systems.

Summary

Anthropic open-sourced 11 knowledge-work plugins that transform Claude into a role-specific specialist for functions like sales, data engineering, legal, and bio-research. Each plugin bundles domain skills, slash commands, MCP tool connectors (Slack, BigQuery, Jira, Snowflake), and sub-agents into a reusable artifact, installable via Claude Cowork or `claude plugin install` from the CLI. The architecture follows a standard directory structure with `.claude-plugin/`, `.mcp.json`, `commands/`, and `skills/`, enabling teams to customize workflows, tooling, and terminology without rebuilding agent logic.

Author

anthropics