@MythOS MCP (@Model Context Protocol) connects your MythOS knowledge library to AI assistants like @Claude, @Cursor, Windsurf, and any MCP-compatible client. It lets you search memos, read content, create and update memos, chat with your library via RAG, explore communities, and sync changes — all from within your AI tools. This is the Memory layer of @The Augmentation Stack in production: structured, persistent, identity-aware knowledge accessible from any AI client on any device.

How Context Loading Works

Every MCP interaction begins with get_context — a tool that loads your Augmentation System memos before any memo is created or updated. This means your AI collaborator knows your formatting rules (Mythos Style), collaboration values (Mythos Soul), who you are (Mythos Human), and what your collaboration has learned (Mythos Memory) before it writes anything. This is what makes MythOS MCP identity-aware — the AI doesn't just access your notes, it knows who you are. See: @Collaborative Augmentation. The four augmentation memos are private, editable memos in your library tagged #mythos-augmentation. You can customize them at Settings > Augmentation or by telling your AI to update them during conversation. When your AI learns something about how you work, it can be stored in your Mythos Memory — and every future session, across every client, starts with that knowledge.

What It Does

MythOS MCP exposes 19 tools to any connected AI assistant:

get_context — Load your augmentation memos (Soul, Style, Human, Memory). Called automatically before creating or updating memos
search_memos — Search your library by query, tags, or visibility (including private memos for the owner)
read_memo — Read the full markdown content of any memo, including collaborator notes
create_memo — Create new memos with title, content, tags, and visibility
update_memo — Update existing memo content and tags
delete_memo — Move a memo to the trash (soft delete, restorable)
list_tags — List all hashtags with usage counts across your library
chat_with_library — Ask questions answered by @RAG over your knowledge base. See: @MCP vs RAG
delta_sync — Get memos that changed since a given timestamp
get_memo_index — Lightweight index of all memos (titles, tags, no content)
list_communities — Browse public MythOS communities
search_community_posts — Search posts within a community
list_tasks / create_task / complete_task / update_task — Manage tasks (standalone and memo-embedded)
read_daily_memo / add_daily_task / add_daily_entry — Interact with daily memos It also exposes 8 resources (memo content, creator profile, memo feed, community metadata, and the four augmentation memos) and 3 prompts (enrich memo, summarize library, find connections). Full tool reference: @MythOS MCP — Tools, Resources & Prompts.

Distribution Channels

Claude.ai / Claude Mobile (OAuth — one-click setup)

MythOS is available as a connector in Claude. When you add it, you'll authorize via OAuth and your library becomes accessible in any Claude conversation — including the iOS and Android mobile apps. See: @MythOS on Claude Mobile.

Go to claude.ai/customize/connectors
Click Add custom connector
Enter: https://mythos.one/api/mcp
Click Connect and authorize access to your MythOS library The OAuth flow uses PKCE with Dynamic Client Registration, JWT access tokens (1-hour TTL), and refresh token rotation (90-day TTL). Scopes: read (search, read, list) and write (create, update memos).

@Claude Code (CLI)

For developers using Claude Code in the terminal:

claude mcp add mythos \
  --scope user \
  https://mythos.one/api/mcp \
  --header "Accept: application/json, text/event-stream" \
  --header "x-mythos-key: YOUR_API_KEY" \
  --header "x-mythos-username: YOUR_USERNAME"

Generate your API key at mythos.one/settings/api-keys (requires Scholar or Oracle tier). See the @MythOS Claude Code Skill for an automated setup script.

Claude Desktop / @Cursor / Windsurf (Npm Package)

For any MCP client that supports stdio transport:

{
  "mcpServers": {
    "mythos": {
      "command": "npx",
      "args": ["-y", "mythos-mcp"],
      "env": {
        "MYTHOS_API_KEY": "YOUR_API_KEY",
        "MYTHOS_USERNAME": "YOUR_USERNAME"
      }
    }
  }
}

The npm package is mythos-mcp (npmjs.com/package/mythos-mcp).

Technical Architecture

Server: Built with @modelcontextprotocol/sdk v1.27.1, Streamable HTTP transport
Auth: OAuth 2.0 (authorization code + PKCE) for Claude connectors, API key auth (x-mythos-key header) for Claude Code and the npm package
Backend: All tool calls route through the MythOS v3 API with per-user scoping
Safety annotations: Every tool has readOnlyHint, destructiveHint, and idempotentHint annotations per MCP spec
Augmentation: get_context loads the creator's augmentation memos at the start of each memo-writing interaction. Managed via the Augmentation System dashboard at settings/augmentation
OAuth endpoints: Discovery at /.well-known/oauth-authorization-server and /.well-known/oauth-protected-resource/api/mcp

Source Code

npm package: github.com/citizens-of-one/mythos-mcp
Claude Code skill: @MythOS Claude Code Skill — github.com/citizens-of-one/mythos-claude-code
MCP server (canonical): studio-mythos-v3/src/mcp/create-server.ts This is the piece that makes @human-AI augmentation portable. Before MythOS MCP, my knowledge system was trapped on my laptop. Now it travels — Claude.ai on the web, Claude on my phone, Claude Code in the terminal, Cursor in the IDE. Same library, same augmentation context, same identity. The AI knows who I am regardless of which client I open. The OAuth flow for Claude.ai was the breakthrough for non-technical users. One click to connect, and suddenly your entire MythOS library is available in every Claude conversation. No API keys, no config files, no terminal. That's the bar for AI memory systems — if it doesn't work on your phone, it's not memory. It's a dev tool.

Contexts

#agentic-augmentation
#ai-integration
#model-context-protocol
#mythos-feature