Twenty specialist legal skills became one working system in a single day
Over 20 specialist Claude skills unified into one legal-AI system covering three domains, from concept to demo in a single day.
This builder took a scattered set of legal tools and turned them into one system that reads a document, works out what kind of problem it is, and drafts the right response. They did it, demo and all, inside a single day. Their story shows how a clear idea and ruthless focus can turn existing work into something whole.
Legal-AI system: detect, route, draft, check
Before
They had already built useful Claude skills for council tax and bailiff enforcement, employment tribunal claims and private parking appeals. But the tools lived apart, each one siloed. They kept seeing the same real-world pattern: one person, several legal pressures at once, and no budget for a separate solicitor for each.
“Identified a recurring real-world pattern — one person, multiple simultaneous legal pressures, no budget for separate solicitors per issue — and built specialist skills for each domain rather than one generalist legal chatbot”
In their own words
The turning point
Facing a genuine same-day deadline, they used ICE prioritisation, weighing impact, clarity and ease, to pick one idea over the rest. They chose to unify what already worked rather than build anything new. That decision gave the work a shape: a single detect, route, draft and check architecture tying every skill together.
After
They now have one working system that coordinates more than 20 specialist skills across three legal domains. They showed it live against three different document types, with a persona, recordings, an architecture diagram and a full write-up. Every document and persona in the demo was anonymised, so no real client data was ever used.
What it lets people do
Someone facing several legal problems at once can bring in a document and have it routed to the right procedure automatically. Each skill carries domain-specific statutory checklists and drafting conventions, so the output follows proper process instead of generic advice.
What they learned
They came to see that the access-to-justice gap is less about legal complexity and more about the cost of a solicitor's time. Narrow, well-scoped skills working together proved more reliable than one broad system, and a hard deadline taught them that cutting scope beats chasing polish.
“Building under a hard deadline forced ruthless prioritisation — cutting scope (one category, existing assets only, no new development) was more valuable than chasing polish”
In their own words
Legal-AI system: detect, route, draft, check was built during AI Mastery.
Read the full submission
Written 9 August 2026 Submitted by the storyteller
Every story is written by a Mindvalley student or by someone our teachers have worked with. How we collect and check stories
More from AI Mastery
