Your AI writes great specs. Now your team's AI can actually read them.

Share requirements, design docs, and architecture decisions. Teammates read them with AI — getting instant answers, reviews, and full project context.

Free · No credit card required

Your AI writes great specs. But sharing them breaks the AI connection.

You generate a spec with AI, then paste it into Slack — losing all context
Teammates copy-paste docs back into their AI for review — manual and fragile
Project knowledge is scattered across Notion, Slack, and Google Docs

AI Chat Demo

See how AI answers questions based on the document content

pastemd.io/s/iphsiu6M...

Payment System v2 — Technical Requirements

Overview

This document outlines the requirements and technical design for migrating our payment processing from Stripe-only to a multi-provider architecture supporting Stripe, PayPal, and regional payment methods.

Generated with: Claude Code from JIRA tickets PM-301 to PM-315 Status: Draft — Awaiting team review

Functional Requirements

  1. Multi-provider support: Process payments via Stripe, PayPal, and regional providers (Toss for KR, Razorpay for IN)
  2. Automatic failover: If primary provider fails, retry with backup provider
  3. Unified webhook handling: Single endpoint normalizing events from all providers
  4. Idempotency: All payment operations must be idempotent with client-generated keys

Non-Functional Requirements

  • Latency: < 3 seconds (p99)
  • Availability: 99.95% uptime
  • PCI DSS Level 1 compliance

API Design

POST /api/v2/payments

ParameterTypeDescription
amountintegerAmount in smallest currency unit
currencystringISO 4217 (USD, KRW, INR)
providerstring"stripe", "paypal", or "auto"
idempotency_keystringClient-generated unique key

Failover Strategy

  1. Route to preferred provider based on currency/region
  2. On 5xx or timeout (>5s), retry with backup provider
  3. Alert on-call if failover rate exceeds 5%...
AI Chat
AI chat will start automatically...

Your AI → Their AI. Full Context Preserved.

No more copy-pasting specs into Slack. Share a link — your team's AI reads it directly.

You + AI

Generate the spec

PasteMD

Share with full context

Team + AI

Review, question, build

Full context preserved at every step. No copy-paste. No context lost.

Connect Your AI in 30 Seconds

One command. Your AI coding assistant can read every shared document.

Terminal
claude mcp add --transport http pastemd https://api.pastemd.io/mcp

Your Entire Project, One AI Context

Group specs, schemas, and runbooks into a project. AI understands the full architecture — ask anything.

Payment System v25 documents
Technical Requirements & API Design
Database Schema & Migrations
Authentication & Security Spec
Error Handling Guidelines
Deployment Runbook
AI understands your entire project

What's the auth flow for the payment API?

AI pulls from your Security Spec and API Design docs together to explain the complete authentication flow — no context switching.

Connected AI Workflow

Your AI generates specs → share a link → teammate's AI reads them. Full context preserved, no copy-paste.

AI-Powered Review

Teammates ask AI about your spec. It explains the design, answers questions, and finds gaps before you write code.

Project Knowledge Base

Group all specs, docs, and decisions into a project. AI understands the full context across every document.

How It Works

From writing to sharing to value — in three simple steps

Step 1

Write in Markdown

Write your content in Markdown — or paste existing docs. No formatting hassle.

Step 2

Share a Link

Get an instant shareable link. Set password, expiration, or keep it simple.

Step 3

Your Team's AI Reviews & Builds

Teammates read your spec with AI — review, question, and start building

From spec to code, with AI at every step. Start sharing now.

Free · No credit card required