Legal AI Policy Skill
Draft a firm AI policy: what is allowed, what is banned, and how output is verified.
Installs into Claude, ChatGPT, Gemini, Copilot, any AI that reads a SKILL.md. Drop the SKILL.md into your AI of choice.
| Category | Governance |
|---|---|
| Price | $99 · or included with membership |
| Works in | Claude, ChatGPT, Gemini, Copilot |
| Format | SKILL.md + references |
A folder your AI loads on demand. It answers from the source, and says [NOT IN SOURCE] instead of inventing.
The method, ready to use.
Generates a firm AI policy covering permitted tools, confidentiality and privilege, the verification step, disclosure to clients and a review cadence — with placeholders for your own decisions.
Is this you?
Draft a firm AI policy: what is allowed, what is banned, and how output is verified.
It triggers on
- draft an AI policy
- what is allowed
- confidentiality rules
- AI governance
What it produces
- review
- drafting
Pairs with: LexGPT Harness, Microsoft 365 Copilot.
A folder your AI loads on demand.
- SKILL.md — the method and the Rules, with every claim labelled authority or reasoning.
- cheatsheet.md — the six questions to answer before sign-off.
- manifest.json — the provenance manifest.
- README.md — how to install it in your AI of choice.
Installs into Claude, ChatGPT, Gemini, Copilot, any AI that reads a SKILL.md.
The model labels every claim [AUTHORITY] or [REASONING]. Anything not in the source is [NOT IN SOURCE]. It never invents a case, a clause or a figure.
From download to answer in three steps.
Download & unzip
You get a folder with SKILL.md, its references and a cheatsheet.
Drop it into your AI
Add SKILL.md to your AI of choice — a Claude Project, a ChatGPT Custom GPT, a Gemini Gem, or your tool's skills folder.
Ask
Ask in this domain. It answers from the source, cites the section, or says [NOT IN SOURCE].
A ledger you can hand to a file.
Whatever you ask, Legal AI Policy Skill returns the work product plus a claim ledger: what was claimed, where it came from, and whether it is authority or reasoning.
- Every claim labelled authority or reasoning.
- Anything not in the source marked
[NOT IN SOURCE]. - A named lawyer signs off — the skill never does.
claim source type --------------------------- ------------------ ----------- limitation period six years Limitation Act s 5 AUTHORITY clause likely unenforceable analysis REASONING "the leading authority..." — NOT IN SOURCE notice period 30 days contract cl 12.3 AUTHORITY
Drop the SKILL.md into your AI of choice.
Claude
Add SKILL.md to a Project's knowledge, or to your skills folder.
ChatGPT
Attach it to a Custom GPT's knowledge and set the instructions.
Gemini
Add it to a Gem's files, or paste it at the top of a session.
Copilot
Reference it in an agent, or paste it into a prompt.
Before you install.
What exactly is a skill?
A skill is a document your AI loads on demand: a SKILL.md with the method, reference files, a cheatsheet and a provenance manifest. The model answers from it and replies [NOT IN SOURCE] instead of inventing.
Which AIs does it work with?
Claude, ChatGPT, Gemini, Copilot and any tool that reads a SKILL.md. Drop the file into your AI of choice — a Claude Project, a ChatGPT Custom GPT, a Gemini Gem, or your tool's skills folder.
Will it answer from the source?
Yes. Every package carries explicit answering rules: cite the section for each claim, label authority vs reasoning, and never invent a figure, citation or rule.
Is it kept current?
Yes. Method and practice skills are reviewed quarterly, and members can re-download the latest version any time from their dashboard.
Legal AI Policy Skill
Draft a firm AI policy: what is allowed, what is banned, and how output is verified.