What is this?
It is an AI agent harness that builds a live, evolving 'biosecurity world' around any target you define, continuously gathering evidence and recommending protective actions.
It is an AI agent harness that builds a live, evolving 'biosecurity world' around any target you define, continuously gathering evidence and recommending protective actions.Is this a graphical app or a command-line tool?It is terminal-first: you control targets, sources, evidence, and simulations from a CLI, with an optional read-only web visualizer for maps and relationships.
It is terminal-first: you control targets, sources, evidence, and simulations from a CLI, with an optional read-only web visualizer for maps and relationships.How do I get it running?You configure your AI agent and tell it what to protect in plain language; it then starts investigating and tracking automatically, and your session's targets and world state persist locally across restarts.
You configure your AI agent and tell it what to protect in plain language; it then starts investigating and tracking automatically, and your session's targets and world state persist locally across restarts.Which AI agents or models does it work with?Any agent or model provider works through an OpenAI-compatible adapter, with native adapters built for Codex and Claude.
Any agent or model provider works through an OpenAI-compatible adapter, with native adapters built for Codex and Claude.Is this meant for pathogen research or clinical diagnosis?No, it is explicitly scoped to defensive biosecurity monitoring and protection, not pathogen engineering or clinical use.
No, it is explicitly scoped to defensive biosecurity monitoring and protection, not pathogen engineering or clinical use.Does my data get uploaded anywhere?No, the agent runs locally and targets, files, and secrets are not uploaded by default; any external content it pulls in is treated as untrusted.
No, the agent runs locally and targets, files, and secrets are not uploaded by default; any external content it pulls in is treated as untrusted.What kinds of targets can I track?People, animals, plants, products, places, organisations, or several connected targets at once, described in plain language.
People, animals, plants, products, places, organisations, or several connected targets at once, described in plain language.What sources does the agent monitor?Official and scientific sources, news and the open web, public OSINT, sensors and surveillance feeds, and any connected data you add.
Official and scientific sources, news and the open web, public OSINT, sensors and surveillance feeds, and any connected data you add.Can it do more than just monitor?Yes, from the current live state it can simulate forward scenarios and recommend defensive actions, with observed, inferred, and simulated claims always kept distinct.
Yes, from the current live state it can simulate forward scenarios and recommend defensive actions, with observed, inferred, and simulated claims always kept distinct.What are similar projects?casbin-gateway, codex-security, claude-code, Raven, evenfire, canonry, ruflo, Shadowbroker
casbin-gateway, codex-security, claude-code, Raven, evenfire, canonry, ruflo, ShadowbrokerWhich topics is it in?fastify, surveillance, osint, harness, cybersecurity, sqlite, anthropic, dashboard, monitoring, claude, web-app, ai-agent
fastify, surveillance, osint, harness, cybersecurity, sqlite, anthropic, dashboard, monitoring, claude, web-app, ai-agent