Continuity Documentation
Install, activate, and operate Continuity. For benchmark results and methodology, see the whitepaper.
Quick Navigation
Getting Started
What is Continuity?
Decision governance for AI-assisted codebases: git-owned `.continuity/`, session handoff, MCP tools, and bounded search when you look up.
In-Loop Retrieval
Decisions surface automatically on file-touching tool calls. See whitepaper §3 for the pattern.
Installation — VS Code extension
- Open VS Code Marketplace (current release: v3.0.156)
- Click Install
- Reload VS Code
Installation — npm CLI (terminal)
Latest: @continuity/cli@3.9.4 (includes MCP via npm dependency — one package).
npm install -g @continuity/cli@^3.9.4
In your project repo — one-command onboarding (MCP, hooks, handoff, recall verify):
continuity onboard continuity hooks install
continuity onboard (npm-only) also wires a shell enter-project hook (status on cd) and a Claude Code SessionStart hook (auto-injects SESSION_HANDOFF.md). Git hooks refresh handoff every commit. Optional: continuity watch for passive refresh; continuity doctor for health checks.
First Launch
- Zero-click setup — auto-detects AI tools
- Initial scan — creates draft decisions from codebase
- Git seeding — pre-populates from commit history
- Creates
.continuity/folder with plain JSON
14-Day Free Trial
Full Pro access, no credit card required. After the trial, decision logging stays free — auto-capture and full handoff require Pro.
Features
14-Day Pro Trial
Auto-capture, full session handoff, unlimited decisions
Auto-detects and configures AI tools
19 detection points auto-capture decisions
Pro Features
UpgradeCurated context at startup
Decisions land while you work
4-phase consolidation
Auto-outdating
Refactoring safety
When you look something up
Explore relationships
No storage limits
License Management
Activation
- Press
Cmd/Ctrl+Shift+P - Type "Continuity: Activate License"
- Enter your license key and email
License Types
Cancel anytime, 14-day free trial
Save 18%, 14-day free trial
Never expires, all future updates
MCP Integration
What is MCP?
Continuity uses the Model Context Protocol to provide AI assistants with persistent memory and decision logging.
Memory Tools
log_decisionLog decisions with question, answer, and tags
search_decisionsKeyword search; semantic ranking when the local index is built
get_quick_contextLoad project history at conversation start
session-handoffComplete project context for handoffs
API Reference
Licensing runs through Lemon Squeezy from the VS Code extension and @continuity/cli. There is no public Hackerware /api/devices API on this site.
POST …/licenses/activateCreate instance on this machine (extension / CLI)
POST …/licenses/validateRefresh entitlement with license_key + instance_id
POST …/licenses/deactivateFree a slot (requires instance_id from that device)
Base: https://api.lemonsqueezy.com/v1 · Customer guide: /license-transfer
Troubleshooting
License activation fails
- Verify license key (check for typos)
- Use the email from your purchase
- Check internet connection
- Maximum activations: deactivate on old machine or email support
MCP not working
- Verify Continuity is installed and activated
- Check MCP server logs in VS Code Output
- Restart your AI assistant
FAQ
How does Continuity differ from CLAUDE.md?
CLAUDE.md loads the full file every conversation. Continuity retrieves only relevant decisions on demand. Benchmark comparison: see whitepaper §4.3.
How does Continuity compare to MemPalace?
In a blind head-to-head on Continuity’s own repository (October 2026, 50 author-written queries; product names hidden from the judge, order randomized and every query judged in both orders), Continuity went 41–5–4, 45–3–2 and 46–0–4 under three LLM judges from different vendors, with mean relevance 0.83 vs 0.47 under the claude-fable-5 judge. It is a home-field comparison, not a general one. See whitepaper §4.3 for the full table and caveats.
Is my data secure?
All decisions stored locally in .continuity/ — never on our servers.
What AI assistants work?
Claude Code, Cursor, GitHub Copilot, Google Gemini, and more via zero-click setup.
Where can I read the methodology?
The Continuity whitepaper covers fixtures, conditions, scoring, and inter-judge validation. Replication artifacts are linked in §A.7.
Still have questions?
Our team is here to help.