LSO compliance panel and AS IS disclaimers added to the UI

Two commits add jurisdiction-specific professional-responsibility framing to the product: an `LSOComplianceTab` component spelling out Law Society of Ontario duties, and "AS IS / not legal advice" disclaimers across the README, chat view, and initial landing screen.

compliancechat-ui

6749e42 adds frontend/src/components/LSOComplianceTab.tsx, a new tab type (kind: "compliance") in AssistantSidePanelTab, and a prompt extension in prompts.py. The compliance tab surfaces three LSO duties by rule number: Duty of Competence (Rule 3.1-2), Duty to Supervise (Rule 6.1-1), and Transparency. It also adds an "Export AI Supervision Log" button. That button currently calls alert() - it's a stub, not a working export.

The prompts.py change extends Rule 2 into a numbered sub-list: use "Blackline" over "Redline," use "Lawyer"/"Counsel" over "Attorney," use Canadian spelling, and adds LSO Rule 3.1-2 and Rule 3.3-1 as explicit constraints on the model (no hallucinated citations, refuse requests involving confidential client data in unsecured prompts).

0398d2c adds the disclaimer surfaces. The README gets a blockquote: "AS IS", without warranty of any kind. Anyone using this software does so entirely at their own risk. ChatView appends a second line below the existing "You are fully accountable for AI-fabricated citations" footer. InitialView replaces the generic "AI can make mistakes" note with a bolded "Open Source Project - Provided 'AS IS'" header.

The compliance tab content itself is the most reusable part. The actual LSO rule text and the "never rely on AI as an oracle / you are accountable for fabricated citations" framing make reasonable reference material if you're building a compliance disclosure panel for Ontario lawyers.

So what The disclaimer copy and `LSOComplianceTab` structure are directly reusable boilerplate for any legal-AI product targeting Ontario. The export-log feature is non-functional. Skip the prompt sub-rule numbering - it creates a formatting inconsistency in the existing prompt structure.

View this fork on GitHub →

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

2 commits from jrklaus8/mike-Canada, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
6749e429 feat: Integrate LSO AI Compliance framework (Competence, Supervision, Transparency) MikeOSS Bot 2026-05-22 ↗ GitHub
0398d2c7 docs: Add open source AS IS and use at your own risk disclaimers MikeOSS Bot 2026-05-22 ↗ GitHub

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-545.md from inside the repo you want the changes in.

⬇ Download capture-thread-545.md