Skip to main content
Glama
XAVIVILLARREAL

mcp-memoria

by XAVIVILLARREAL
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

mcp-memoria

MCP server: capa unificada de memoria para agentes IA de desarrollo.

Un solo endpoint MCP que expone 13 tools combinando:

  • codebase-memory-mcp — grafo de código (search, trace, query, detect)

  • Engram — memoria episódica persistente (save, search, context)

  • orient — briefing warm-start (V3Code-inspired)

  • AI-CONTEXT.md — snapshot destilado de arquitectura

Compatibilidad

Cualquier herramienta que hable MCP:

  • OpenCode

  • Claude Code

  • Codex CLI

  • Cursor

  • Aider

  • Gemini CLI

  • KiloCode

  • Kiro

Related MCP server: repowise

Instalación

npm install -g mcp-memoria

Uso

Como MCP server (stdio)

# OpenCode — agregar a opencode.json:
{
  "mcp": {
    "mcp-memoria": {
      "enabled": true,
      "type": "local",
      "command": ["mcp-memoria"]
    }
  }
}

CLI directo

mcp-memoria --tool=orient     # Briefing warm-start
mcp-memoria --tool=ai_context  # Leer AI-CONTEXT.md

Tools

Tool

Fuente

Descripción

orient

scripts/orient.sh

Briefing warm-start: posición, paridad, grafo, distillate, memoria

ai_context

AI-CONTEXT.md

Snapshot destilado: rutas API, hotspots, tablas BD, apps

code_search

codebase-memory

Búsqueda BM25/semántica en grafo de código

code_trace

codebase-memory

Trazado de llamadas (callers/callees/impacto)

code_graph_query

codebase-memory

Queries Cypher sobre el grafo

code_read

codebase-memory

Leer código fuente de una función/símbolo

code_changes

codebase-memory

Detectar cambios recientes y su impacto

code_architecture

codebase-memory

Overview de arquitectura (packages, clusters)

mem_save

Engram

Guardar observación persistente

mem_search

Engram

Buscar en memoria persistente

mem_context

Engram

Contexto reciente de sesiones anteriores

mem_stats

Engram

Estadísticas del sistema de memoria

mem_reindex

Ambos

Re-indexar grafo + regenerar AI-CONTEXT.md

Detección automática de proyecto

El server auto-detecta la raíz del proyecto buscando:

  1. CLI arg --project-root=/path

  2. Env var PROJECT_ROOT

  3. Walk-up desde cwd buscando AGENTS.md, docker-compose.yml, turbo.json

  4. Paths conocidos (/config/ERP-DOCKER-STACK)

Licencia

MIT

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • repowise

    repowise-dev
    A
    license
    A
    quality
    A
    maintenance
    Provides AI coding agents with five intelligence layers (dependency graph, git history, documentation, architectural decisions, code health) via nine MCP tools, enabling deep codebase understanding and reducing exploration cost.
    10
    6,415
    AGPL 3.0

  • mnemo

    arttttt
    A
    license
    Not graded
    quality
    A
    maintenance
    A persistent, local memory layer for AI coding agents that remembers decisions, bugs, and rules across sessions with three core MCP verbs (recall, remember, search).
    Apache 2.0

  • Tsumugi

    archfill
    A
    license
    Not graded
    quality
    A
    maintenance
    An MCP-based memory layer for autonomous agents that provides observation, memory recall, and dreaming capabilities. It enables AI clients like Claude Code and Codex to store and search structured memories via a unified interface.
    180 npm
    Apache 2.0