Anthropic has released a comprehensive repository of reference agents, skills, and data connectors designed to automate complex financial services workflows, including investment banking, equity research, private equity, and wealth management. The system is architected to function either as a Claude Cowork plugin or via the Claude Managed Agents API for custom deployment. Technical implementation utilizes a modular structure where "vertical plugins" contain core skills, slash commands (such as /comps, /dcf, and /earnings), and Model Context Protocol (MCP) connectors. These connectors enable seamless integration with high-fidelity data providers including S&P Global, LSEG, Bloomberg, FactSet, Morningstar, and PitchBook. The repository provides specialized agents—such as the Pitch Agent, Market Researcher, and GL Reconciler—which bundle specific skills and system prompts to handle end-to-end tasks like drafting CIMs, updating financial models in Excel, or performing KYC screening. For enterprise environments, the repository includes admin tooling for provisioning Claude within Microsoft 365 via Azure, Vertex AI, or Bedrock. Developers can extend the framework by adding new skills in markdown/YAML and using provided Python scripts like `sync-agent-skills.py` and `check.py` to validate manifests and propagate skill updates across the agent ecosystem.
Read original
github-trending/python