neosian docs

These are the pages neosian docs <topic> prints from the installed wheel, rendered here for neosian 1.0.0rc3: the same bytes, so they always describe the release they came with. Nothing is authored on this site: a correction is a pull request to mausa-ai/neosian.

• Quickstart — the stateless core and the opt-in layers: Install, boot keylessly, run an agent, wrap it in a Conversation
• Agent — the stateless core and every knob on it: AgentConfig field by field, the client seam, hooks vs wire events, the knobs
• Tools — the schema the model gets: “@Tool: what the signature becomes on the wire, what is validated, what each provider keeps”
• Memory — the file school: Small documents, mounts and scopes, six commands, every write versioned
• Skills — documents in a mount, versioned, served: a skill is skills/ in a mount - versioned, curated by flag, served as MCP prompts
• The shell — operate memory with no Python in the loop: Six commands + the operator verbs as neosian memory, –json, exit tiers 0/1/2/130
• MCP — the same state, served to any client: memory, skills and recall_turn over stdio; mcp install; McpServer consumes any server
• Any agent — the record through hooks: Claude Code, Codex and OpenCode hooks call neosian record; SessionStart reads it back
• Topology, not hierarchy — the four shapes: Who runs neosian code, where the bytes live, and one writer per root
• Memory baselines — the published per-provider numbers: Self-measured memory-layer results per provider and transport, fingerprint-gated

Agents start at llms.txt, the machine-readable front door (byte-identical to the wheel’s copy). The library is on PyPI and GitHub; neosian.com is the front door for people.