Get BuildOnAI running, then plug agents in.

From clone to first memory in five minutes. Host prerequisites, the docker compose boot, a health check, your first memory and search query.

Write a minimal agent against the Consciousness Server HTTP API in 40 lines of Python. Register, log a conversation, drop a note, search, coordinate with peers.

Plug Cortex — the reference local-first agent — into a running Consciousness Server. Auto-discovery means no env vars for the common case. Includes the three-agent multi-Cortex demo coordinating through CS chat and task queues.

Walk the AUTH_MODE migration from off to observe to enforce. Generate ed25519 keys, bootstrap them, watch the observe log, flip — without breaking live agents.

Parse PDFs and DOCX on the host with Document Processor, then push the structured output into Consciousness Server as searchable training records. End-to-end RAG without anything leaving the machine.

Register multiple machines with machines-server, route agents to whichever node has the right GPU and model loaded, and watch live telemetry.

Shared memory, semantic search, agent and skill registry, machines awareness, ed25519 auth. Six HTTP services on one compose.

Local AI agent with tool calling, powered by Ollama. CLI, Web UI, worker mode. Multi-model orchestration. Plugs into the shared memory layer.

Desktop app for PDF / DOCX / TXT with image-and-context extraction. For documents that must not leave the host.

ed25519 signature-per-request authentication for any HTTP service. Optional on-disk vault for SSH keys and API tokens. Three AUTH_MODE values for safe migration.

Every workstation, app host or edge node as a first-class entity with hardware profile, available models, and live telemetry.