Roadmap

Six weeks to a live demo, then four milestones

Demo scenario: an injection hidden in a repository README makes Claude Code read credentials and send them out. Live, the action is blocked and a decoy fires.

Roadmap
Demo

The six-week demo

Weeks 1 to 6: from hooks to three installed teams

Weeks 1–3: Claude Code hook integration and a full action log, plus three core policies: block credential reads, block unknown domains, approve deletes. Weeks 3–6: a decoy layer with fake AWS keys and a honeypot MCP tool, a one-page console with live actions and alerts, then the demo script, a recording and installs at three design-partner teams.

M1 · Observe

Hook integration, a full action log and replay across sessions.

M2 · Control

Policy engine across files, commands, network egress and credentials, with approvals.

M3 · Decoys and code audit

A complete decoy system, plus AST rules and an LLM review pipeline that audits agent-written code.

M4 · Ready to sell

Team console and self-hosted deployment for mid-size engineering teams.

Platform

Beyond M4

The full platform

Network gateway and deception proxy, real-time threat analysis and automated response, self-protection and the hardened zero-trust workspace.

Funding path

The demo is the basis for design partners and a pre-seed round. Without outside funding, paid pilots fund the team.

Design partner program

Install the demo for free and shape the product

Engineering teams using AI coding agents can run the six-week demo at no cost: open-source client, an action log that never leaves your machine, three default policies, decoys and a one-page console.