How Continuity Works

Install the extension, log your first few decisions, and give your AI coding assistant synthetic memory. Start free, upgrade when you need more.

Step 1

Start logging your decisions

Hit Cmd+Shift+L and answer simple questions: "Why did I choose X over Y?" or "Why implement this approach?" Your first 5 decisions are free—enough to see if AI actually remembers.

  • 5 free decisions to start—no email required
  • Pick from 5 templates or write your own
  • Tags auto-extracted, duplicates filtered
  • Want more? Share your email → get 10 more decisions + 7-day Pro trial

# Your first 5 decisions (free)

[You press Cmd+Shift+L]

Pick a template:
• Why choose X over Y?
• Why implement this approach?
• Why refactor this code?
• Why use this library?
• Custom question

Example decision:
Q: Why use Zod over Joi for validation?
A: Zod gives us TypeScript-first design, better
   type inference, and smaller bundle size.

✅ Logged instantly
✅ Tags auto-extracted: [validation, typescript, zod]
✅ Available to AI in next session

After 5 decisions → share email for 10 more + Pro trial

# Where decisions live

.continuity/
├── decisions.json          # All your decisions
├── SESSION_HANDOFF.md      # Auto-generated AI context
└── embeddings.json         # Semantic search (Pro only)

# What a decision looks like:
{
  "id": "1761121236686-bgocje",
  "question": "Why use Zod over Joi?",
  "answer": "Zod for TypeScript-first design, better
            type inference, smaller bundle.",
  "tags": ["validation", "typescript", "zod"],
  "timestamp": "2025-10-22T08:20:36.686Z"
}

✅ Plain JSON you can read and edit
✅ Commit to git with your code
✅ Never sent to the cloud
✅ SESSION_HANDOFF.md updates automatically
Step 2

Everything stays in your project

Decisions go in a .continuity/ folder as plain JSON. No cloud sync, no vendor lock-in. Commit to git, grep them, or edit them manually. You own the data.

  • Readable JSON—no database, no proprietary format
  • SESSION_HANDOFF.md auto-generates with architecture, commits, and decisions
  • Commit to git and share with your team
  • Pro users get semantic search to find decisions faster
Step 3

Synthetic memory loads automatically

Open Claude, Cursor, or any AI tool. Your decisions load automatically. Ask "Why did we choose X?" and it knows. No copy-paste, no re-explaining.

  • Works with Claude Desktop, Claude Code, Cursor, Cline, Roo Code
  • AI can log new decisions and search past ones
  • Fast loading with no noticeable performance impact
  • Uses MCP protocol (Model Context Protocol)—standard across AI tools

# How AI sees your project

[You open Claude Desktop or Cursor]

Continuity loads automatically:
✅ Recent commits
✅ Project architecture
✅ All your decisions
✅ Current git branch

Real conversation:
────────────────────────────────────────────────
You: "How should we validate API inputs?"

AI:  "You're using Zod for validation (logged
      3 days ago in decision #47). Here's a
      schema that matches your existing style..."
────────────────────────────────────────────────

AI can also:
• Log new decisions for you
• Search past decisions
• Track what you've documented

No re-explaining. No context loss. It just knows.

# Quick verification test

You: "Do you have access to Continuity? What tools
      do you have?"

AI:  "Yes, I have Continuity installed. I can see:

      Available tools:
      • log_decision(question, answer, tags)
      • search_decisions(query)
      • get_compliance_metrics()

      Loaded context:
      • 12 decisions logged
      • Last updated: 2 minutes ago
      • Project: your-app-name
      • Tech stack: Next.js, TypeScript, Tailwind"

✅ Working perfectly!

If AI says "I don't have those tools":
→ Reload VS Code (Cmd+Shift+P → Reload Window)
→ Check Output panel (View → Output → Continuity)
→ Make sure you opened a folder (not just a file)
Step 4

Test it in 30 seconds

Ask your AI: "Do you have access to Continuity?" It'll list the tools it has and show you what context loaded. Instant proof it's working.

  • AI confirms it can see your decisions
  • Shows loaded tools: log_decision, search_decisions, get_compliance_metrics
  • Displays how many decisions it found
  • If it doesn't work, reload VS Code or check the output panel

Works with Your Favorite AI Tools

Cursor IDE

Stop Cursor from forgetting your architecture every session. Synthetic memory loads automatically via MCP.

  • Auto-loads on every Cursor session
  • Works in both chat and composer
  • No .cursorrules needed—Continuity is better

Claude Desktop

Claude Desktop forgets past conversations. Continuity gives it synthetic memory across all sessions.

  • Remembers decisions from last week
  • Works with Claude 3.5 Sonnet
  • One-time MCP configuration

Claude Code (CLI)

Terminal-based Claude with full Continuity integration. Perfect for developers who live in the command line.

  • Fastest setup (automatic)
  • Built-in MCP support
  • Zero configuration required

Cline (VS Code)

Cline users get automatic context loading. No more manually pasting architectural notes into chat.

  • Works in VS Code sidebar
  • Shares same .continuity folder
  • Multi-file edits remember context

Roo Code

Roo Code + Continuity = powerful pair programming with memory. Your decisions persist across refactoring sessions.

  • MCP support built-in
  • Perfect for large refactors
  • Team decision sharing via git

GitHub Copilot Chat

Enhance Copilot Chat with architectural context. Your decisions guide Copilot's suggestions automatically.

  • Better code suggestions
  • Respects your tech choices
  • Works in VS Code chat panel

One extension. All your AI tools share the same synthetic memory. That's the power of the Model Context Protocol (MCP).

Ready to Unlock More?

Start with 10 free decisions. When you need unlimited decisions, advanced analytics, or team features, upgrade to Pro or Ultimate.

Pro

$139

  • Unlimited decisions
  • 3 device activations
  • Semantic search
  • Network graph

Perfect for solo developers

NO LIMITS

Ultimate

Contact Sales

Enterprise pricing

  • 10 device activations
  • Full commercial license
  • Advanced analytics + API
  • Team decision sharing
  • + Everything in Pro

For teams & freelancers

Monthly

$15/mo

  • All Pro features
  • 3 device activations
  • Cancel anytime
  • Try before committing

Flexible month-to-month