finds.dev← search

// the find

zubair-trabzada/ai-legal-claude

★ 1,451 · Python · updated Mar 2026

AI Legal Assistant skill for Claude Code. Contract review, risk analysis, NDA generation, compliance auditing, negotiation strategy, and PDF reports — 14 skills, 5 parallel agents. If you want to learn how to sell this to real businesses, check out the Skool community

A collection of Claude Code slash-command prompts for legal document tasks — contract review, NDA generation, compliance checking, and PDF reports. The '5 parallel agents' are just markdown files that Claude reads sequentially or in parallel depending on how Claude Code handles subagents. This is prompt engineering packaged as a product, not a legal tech tool.

The skill decomposition is sensible — splitting clause analysis, risk scoring, compliance, and recommendations into separate agent prompts means each prompt stays focused and gets better output than one giant prompt. The breadth of coverage (freelancer-specific review, contract comparison, missing-clause detection) addresses real gaps that generic ChatGPT prompting misses. The ReportLab PDF generation script is a practical touch that makes output actually deliverable to a client. Install/uninstall scripts that just copy markdown files to ~/.claude/commands/ are appropriately simple.

This is markdown files, not code — the stars/forks numbers reflect viral README marketing more than engineering merit. There is no actual legal knowledge baked in beyond what Claude already knows; the 'Contract Safety Score 0-100' is Claude making up a number, not a calibrated system. The README prominently pitches a paid Skool community for 'selling this to real businesses,' which is the actual product here; the GitHub repo is lead generation. The disclaimer that this 'does not provide legal advice' directly contradicts the use cases section telling you to charge clients $500-$1,500 per review — that liability exposure is real and unaddressed.

View on GitHub → Homepage ↗

// want more like this?

We dig through GitHub every week and send a few repos picked for what you actually care about — each with an honest take like this one.

Get finds in your inbox → Search again →