$ show anthropics/claude-for-legal
This repository is a comprehensive plugin marketplace and reference implementation for legal practice automation using Anthropic's Claude AI. It contains 12 first-party plugins (commercial, corporate, employment, privacy, product, regulatory, AI governance, IP, litigation, legal clinic, law student, and a builder hub) plus one vendor plugin from Thomson Reuters. Each plugin provides a set of 'skills' (prompt-engineered workflows) invoked via slash commands, a cold-start interview to configure the AI to a specific legal professional's role and preferences, and optional scheduled agents. The repo also includes five managed-agent cookbooks demonstrating multi-agent orchestration with tool isolation (read-only subagents, write leaf, orchestrator with no write access). While the vast majority of content is markdown prompt files, the repository also contains reusable Python and shell scripts for deployment, validation, and linting of agent configurations. The system is designed to be deployed on Claude Code or Claude Cowork, with MCP connectors for legal research tools, document management, and other integrations. It emphasizes citation provenance, injection defense, and consequential-action gates for non-lawyer users.
The LLM identified 3 files worth highlighting for reuse:
A self-contained bash script that deploys managed agents from YAML manifests. It resolves conveniences like system.file and skills.from_plugin, uploads skills to the /v1/skills API, and POSTs agent definitions to /v1/agents. Demonstrates a reusable pattern for programmatic deployment of multi-agent systems with dependency resolution.
A Python script that enforces tool-scope isolation in managed agent configurations. It checks that orchestrators have no write tools, subagents have appropriate restrictions, and that tool sets match declared roles. This is a reusable pattern for enforcing security boundaries in multi-agent systems, preventing privilege escalation through tool misuse.
A Python script that validates subagent outputs against JSON schemas defined in agent YAML manifests. It ensures that data passed between agents conforms to expected formats, catching schema violations before they propagate. This pattern is reusable for any multi-agent pipeline requiring structured data exchange.