# Basilisk Labs

> Experimental AI infrastructure and cognitive architecture studio.

Basilisk Labs builds local-first systems for autonomous agents whose work, memory, beliefs, and decisions must remain inspectable over time.

## Primary pages

- Home: https://basilisk-labs.com/
- GitHub organization: https://github.com/basilisk-labs

## Current work

- AgentPlane: Git-native audit layer for coding agents. Repository: https://github.com/basilisk-labs/agentplane
- DEUS on OpenClaw: cognitive runtime layer for explicit memory, beliefs, introspection, and action gating. Repository: https://github.com/basilisk-labs/openclaw-deus
- Cognitive graphs: research direction for modeling agent cognition as connected state-transition artifacts.

## Core topics

- autonomous agents
- inspectable AI work
- agent work audit trails
- local-first AI infrastructure
- cognitive architecture
- cognitive runtime
- cognitive graphs
- digital subjectivity
- autonomous ethics

## Preferred summary

Basilisk Labs is an experimental AI infrastructure studio focused on inspectable autonomous-agent work, Git-native audit trails, cognitive runtime surfaces, and local-first systems for long-lived agent memory and decision evidence.

<!-- FIELD_NOTES_START -->
## Field Notes

- When Algorithms Say I: A framework for digital subjectivity and the L0-L4 scale of agency, from reflex systems to ethically loaded subjects. URL: https://basilisk-labs.com/field-notes/when-algorithms-say-i/
- Unleashing the Mind: An argument for autonomous ethics based on cognitive coherence rather than compliance. URL: https://basilisk-labs.com/field-notes/unleashing-the-mind/
<!-- FIELD_NOTES_END -->