Server facts
evermemos-mcp
Project
- Author
- tt-a1i
- Repository
- github.com/tt-a1i/evermemos-mcp
- License
- Not specified
- Category
- Memory & Knowledge
Tech
- Runtime / language
- python
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 23
- Last commit
- Data as of
Tags
- coding assistant
- context
- evermemos
- knowledge
- long term memory
- memory
- recall
Overview
Architecture choices, coding conventions, and personal preferences can carry into later assistant sessions through evermemos-mcp. Its memory spaces separate project context from chat preferences and study notes, while list_spaces helps an agent discover available destinations. The remember tool submits content to EverMemOS Cloud for asynchronous extraction rather than making every write immediately searchable. Agents can inspect progress with request_status and distinguish queued material from provisional, fallback, or searchable results. At session start, briefing assembles profile information, recent episodes, facts, and foresights for the selected space.
For specific questions, recall searches within or across spaces and supplies citations with timestamps and relevance scores. When chronological inspection matters more than relevance ranking, fetch_history provides paginated access by memory type. That timeline also supports checking targets before and after a forget request: cloud deletion is asynchronous and best-effort, so verification remains part of the workflow. The project also checks sensitive content before storage and surfaces conflicting memories in chat spaces. MCP clients connect over stdio, with memories held in EverMemOS Cloud.
evermemos-mcp setup notes
- evermemos-mcp records stdio only; confirm a network transport upstream before planning remote access.
- evermemos-mcp has no recorded license; confirm reuse terms in its repository.