evolsb/claude-legal-skill

AI-powered contract review skill with CUAD risk detection, market benchmarks, and lawyer-ready redlines. Works with Claude Code, Codex, Cursor, and 26+ tools.

What it solves

This project provides an AI-powered contract review system designed to be a drop-in skill for AI coding assistants. It addresses the gap between expensive enterprise legal AI and generic checklists by offering position-aware risk detection, market benchmarks, and specific redline suggestions for various contract types (NDAs, SaaS, M&A, etc.) directly within a developer's workflow.

How it works

The tool is implemented as an "Agent Skill" compatible with tools like Claude Code, Cursor, and GitHub Copilot. It leverages the CUAD dataset (containing 41 legal risk categories from real contracts) to analyze documents. It allows users to specify their role in the contract (e.g., buyer vs. seller) to adjust risk flagging and outputs structured JSON redlines that can be converted into professional Word documents or PDFs using a companion tool called legal-redline-tools.

Who it’s for

It is primarily for individuals and developers who need a first-pass legal review of contracts without immediate access to enterprise legal software or a lawyer, allowing them to identify red flags and prepare for negotiations.

Highlights

  • Position-Aware Analysis: Adjusts risk assessments based on whether the user is the customer, vendor, buyer, or seller.
  • Document-Specific Checklists: Tailored reviews for NDAs, SaaS/MSA, Payment/Merchant, M&A, and Broker agreements.
  • Market Benchmarking: Compares contract terms against industry norms to identify "Yellow" or "Red" flags.
  • Concrete Redlines: Provides actual replacement language for clauses rather than generic advice to negotiate.
  • Jurisdiction Awareness: Flags enforceability issues based on specific US state laws (e.g., non-competes in California).

Related

  • Project
    hanlulong/econ-writing-skill

    An Agent Skill that transforms AI assistants into expert economics paper writers by synthesizing best practices from over 50 authoritative guides by leading economists.

  • Project
    internet-court/internet-court-skill

    An open skill for agent-to-agent contracts that unifies payment, negotiation, and dispute resolution protocols to enable trustworthy autonomous commerce.

  • Project
    HoangNguyen0403/agent-skills-standard

    Agent Skills Standard (AGS) is an NPM‑distributed CLI that lets teams publish their coding‑standards, security rules, and SDLC workflow templates as tiny markdown “skills”. The CLI (`ags init` / `ags sync`) creates a `.skillsrc` config, downloads language‑specific skill packs (280 skills across 20+ stacks), and writes a hierarchical router (`AGENTS.md → _INDEX.md → SKILL.md`). AI coding assistants (Cursor, Claude Code, Copilot, Gemini, etc.) read this router and load only the relevant rules for the file they are editing, saving ~85 % of prompt tokens. An optional MCP server can enforce hooks, block secret‑leak edits, and provide runtime audit logs. The project is a genuine AI‑coding‑assistant utility, not a generic tutorial or list.

  • Project
    anthropics/claude-for-legal

    A collection of AI agents and plugins for legal professionals that automates workflows across commercial, corporate, privacy, and litigation practices.

  • Project
    sickn33/agentic-awesome-skills

    AAS Core (Agentic Awesome Skills) is a preview‑stage, open‑source toolkit that lets LLM agents (Codex, Claude, etc.) search a local catalog of 2 115+ reusable instruction “skills”, pick exact IDs for a project, and produce a validated, immutable manifest (`aas‑stack.json`). The core is read‑only: it validates IDs, does not rank or execute them. A CLI can preview a plan, audit risk, and dry‑run a direct installer that copies selected skill directories into `~/.agents/skills`. Specialized plugins bundle related skills for domains like web‑app building or security. The system is designed for auditability and offline control, with safety warnings that it does not certify semantic suitability and that apply/recovery features are experimental.