Open source

A shared memory, attribution and governance layer for AI coding agents — a substrate, not an orchestrator.

• Governed AI Systems
• Evidence, Audit and Replay

A one-click remote dev/ops platform: browser VS Code, multiple AI CLIs and the Octopus Blackboard.

• Governed AI Systems
• Sovereign and Local AI

A standalone governed execution runtime taking a trigger to a result across autonomy, policy and approval boundaries.

• Governed AI Systems
• Responsible AI

The shared Evidence primitive for the Octopus stack — canonical, hashable, tamper-evident records.

• Evidence, Audit and Replay

The governed work-state layer: a task is treated as a verifiable provenance chain, not a to-do item.

• Evidence, Audit and Replay
• Governed AI Systems

A determinism harness for governed event systems, aiming to make every incident byte-for-byte reproducible.

• Evidence, Audit and Replay
• Model Evaluation

Project Memory — organisational, causal memory for AI teams, designed to answer "why", not just "what".

• Evidence, Audit and Replay
• Graph Reasoning

Standalone, governed, auditable intake and normalisation of raw external events.

• Evidence, Audit and Replay
• Privacy-Preserving AI

A governed, auditable, AI-native web and PDF ingestion engine.

• Privacy-Preserving AI
• Governed AI Systems

A static governance linter for AI agent workflows — catch governance gaps before production does.

• Governed AI Systems
• Responsible AI

Governed marketing over MCP with a draft → review → publish flow keeping a human in control.

• Governed AI Systems

Reef — a governed agentic engineering workspace built on an open, evidence-backed approach.

• Governed AI Systems
• Evidence, Audit and Replay

Public-beta editor distribution and payload packaging for Reef. Licence not yet confirmed on GitHub.