<!--
  Full-page Markdown export (rendered HTML → GFM).
  Source: https://neotoma.io/ru/docs
  Generated: 2026-04-26T19:09:31.978Z
-->
# Документация

Вся документация Neotoma, организованная по категориям. Начните с руководства по началу работы, если вы новичок, или перейдите в раздел справки для сведений об API, MCP и CLI.

## Начало работы

-   [
    
    EvaluateHave your agent read this page to decide whether Neotoma fits your workflow
    
    
    
    
    
    
    
    ](/ru/evaluate)
-   [
    
    InstallInstall and initialize Neotoma locally
    
    
    
    
    
    
    
    ](/ru/install)
-   [
    
    WalkthroughEnd-to-end example across operating, building, and debugging
    
    
    
    
    
    
    
    ](/ru/walkthrough)

## Интеграции

-   [
    
    Claude CodePersistent memory for Claude Code's local CLI agent
    
    
    
    
    
    
    
    ](/ru/neotoma-with-claude-code)
-   [
    
    ClaudeStructured state alongside Claude platform memory
    
    
    
    
    
    
    
    ](/ru/neotoma-with-claude)
-   [
    
    ChatGPTDeterministic memory for ChatGPT conversations
    
    
    
    
    
    
    
    ](/ru/neotoma-with-chatgpt)
-   [
    
    CodexCross-task memory and CLI fallback
    
    
    
    
    
    
    
    ](/ru/neotoma-with-codex)
-   [
    
    CursorPersistent memory alongside Cursor context
    
    
    
    
    
    
    
    ](/ru/neotoma-with-cursor)
-   [
    
    OpenClawUser-owned memory for OpenClaw agents
    
    
    
    
    
    
    
    ](/ru/neotoma-with-openclaw)

## Справка

-   [
    
    InstallInstall and initialize Neotoma locally
    
    
    
    
    
    
    
    ](/ru/install)
-   [
    
    REST APIOpenAPI endpoints and parameters
    
    
    
    
    
    
    
    ](/ru/api)
-   [
    
    MCP serverModel Context Protocol actions
    
    
    
    
    
    
    
    ](/ru/mcp)
-   [
    
    CLICommands, flags, and REPL
    
    
    
    
    
    
    
    ](/ru/cli)
-   [
    
    Memory guaranteesAll memory properties on one page
    
    
    
    
    
    
    
    ](/ru/memory-guarantees)
-   [
    
    Memory modelsPlatform, retrieval, file-based, and deterministic memory compared
    
    
    
    
    
    
    
    ](/ru/memory-models)
-   [
    
    FoundationsPrivacy-first architecture and cross-platform design
    
    
    
    
    
    
    
    ](/ru/foundations)
-   [
    
    Agent instructionsMandatory behavioral rules for agents using Neotoma
    
    
    
    
    
    
    
    ](/ru/agent-instructions)
-   [
    
    ArchitectureState flow, guarantees, and principles
    
    
    
    
    
    
    
    ](/ru/architecture)
-   [
    
    TerminologyGlossary of key concepts
    
    
    
    
    
    
    
    ](/ru/terminology)
-   [
    
    Schema managementDefine, inspect, and evolve schema constraints safely
    
    
    
    
    
    
    
    ](/ru/schema-management)
-   [
    
    TroubleshootingCommon failure modes and practical fixes
    
    
    
    
    
    
    
    ](/ru/troubleshooting)
-   [
    
    ChangelogRelease history and documentation updates
    
    
    
    
    
    
    
    ](/ru/changelog)
-   [
    
    All pages (Markdown)Every indexable route as Markdown (SEO summaries)
    
    
    
    
    
    
    
    ](/ru/site-markdown)

## Where the tax shows up

-   [
    
    Context janitorYou re-explain context every session. State that persists across tools.
    
    
    
    
    
    
    
    ](/ru/operating)
-   [
    
    Inference varianceCorrections don’t stick. Deterministic state for agent pipelines.
    
    
    
    
    
    
    
    ](/ru/building-pipelines)
-   [
    
    Log archaeologySame inputs, different state. Replayable timelines and diffs.
    
    
    
    
    
    
    
    ](/ru/debugging-infrastructure)

## Сравнение

-   [
    
    Build vs buyWhen to adopt a state-integrity layer instead of building around observability alone
    
    
    
    
    
    
    
    ](/ru/build-vs-buy)
-   [
    
    Neotoma vs platform memoryConvenience inside one AI product versus portable, auditable state across tools
    
    
    
    
    
    
    
    ](/ru/neotoma-vs-platform-memory)
-   [
    
    Neotoma vs Mem0Retrieval memory for prompt augmentation versus deterministic entity state
    
    
    
    
    
    
    
    ](/ru/neotoma-vs-mem0)
-   [
    
    Neotoma vs ZepKnowledge-graph retrieval versus versioned, schema-bound state
    
    
    
    
    
    
    
    ](/ru/neotoma-vs-zep)
-   [
    
    Neotoma vs RAGRelevant chunk retrieval versus exact state reconstruction
    
    
    
    
    
    
    
    ](/ru/neotoma-vs-rag)
-   [
    
    Neotoma vs file-based memoryMarkdown and JSON portability versus structured guarantees and provenance
    
    
    
    
    
    
    
    ](/ru/neotoma-vs-files)
-   [
    
    Neotoma vs database memoryCRUD rows versus append-only observations and deterministic reducers
    
    
    
    
    
    
    
    ](/ru/neotoma-vs-database)

## Внешние

-   [
    
    GitHub repositorySource code, README, and issues
    
    
    
    
    
    
    
    ](https://github.com/markmhendrickson/neotoma)
-   [
    
    npm packageInstall via npm
    
    
    
    
    
    
    
    ](https://www.npmjs.com/package/neotoma)