Synaplan
Open-source AI platform for companies: self-hosted AI agents, GDPR-compliant chatbot widget, RAG knowledge base, WhatsApp and Outlook AI, MCP, multi-model routing. Docker, Kubernetes or hosted in Germ
Your agreements, organized. An AI agent that files your contracts, notices what's coming due, and helps you take the next step.
BusinessJavaScript · HTMLGitHub @b-i-r-d-oOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
AI agent that reads your personal contracts (leases, mortgages, phone plans, warranties) and flags renewals and deadlines before you miss them. Built for Devpost 'Build with AI: BASICS' Hackathon
Taken from the repository README
Loading the full README.
Read the full README ↗You've signed dozens of contracts you'll never reread. Fine Print reads them for you.
Fine Print is a phone app with an AI agent for your personal agreements: leases, mortgages, phone plans, warranties, anything with terms and conditions. It files your agreements, notices what's coming due, asks you plain questions about your situation, and helps you take the next small step, like drafting a move-out notice.
Created by Joann Dobish for Devpost's Build With AI: Basics hackathon, October 2026.
Not legal advice. Fine Print is a comprehension and guidance tool. It explains public rules in plain language and drafts standard documents for you to review. It doesn't replace a lawyer.
date-party-purpose.pdf) in categories: Housing, Medical, Financial, Marriage, Estate, Work and Other.The sign-in screen links to three fictional sample agreements. You can also find them in samples/:
sample-alberta-lease.pdf: a one-year Alberta lease that has ended. The landlord's address is deliberately missing.sample-mortgage.pdf: a signed home mortgage.sample-employment-contract.pdf: a two-page job offer, good for trying the camera (photograph both pages).Every name, company, address and number in them is invented.
Phone (the app) ──► Helper on Vercel (/api/agent) ──► Claude (Anthropic API)
screens, themes, adds the agent's instructions, reads the contract once,
notice + PDF, holds the hidden API key, answers in a fixed format
saved on the phone checks the reply's format the app can act on
The AI decides what to say; the app decides what's allowed. Legal rules (Alberta's notice rule) are written into the agent's instructions word for word, and dates are calculated in code.
public/: the app (plain HTML, CSS and JavaScript, no framework)api/agent.js: the helper (a Vercel Function)lib/ai.js: the only file that knows which AI is used, so it can be swapped (e.g. for Gemini, or a user's own model)lib/prompts.js: the agent's instructions and the Alberta ruledevpost/: the planning documents (scope, product requirements, technical spec, build checklist)This is a hackathon demo shell, shown with fictional sample documents. An uploaded agreement is sent to one hosted AI (Claude) through the helper to be read. Nothing is stored on the server, and your agreements and chats stay on your phone.
Left out on purpose, and next to build:
node dev-server.js
# open http://localhost:3000
Without an API key it runs in clearly labelled practice mode with canned replies.
ANTHROPIC_API_KEY (Project → Settings → Environment Variables). Never commit a key; .env.example shows the name only.MIT. Fonts: Figtree, Lora and Fredoka, SIL Open Font License (see public/fonts/). PDF: jsPDF 2.5.1, MIT; PDF.js 3.11.174, Apache 2.0 (see public/vendor/).