Skip to main content
Glama
Sign In
deenesjakoruzh
ctx-memory

ctx-memory

by GhadiSaab
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

ctx-memory

Memoria persistente para agentes de codificación LLM: Claude Code, Codex, Gemini CLI, OpenCode.

Cuando inicias una nueva sesión, el agente ya sabe en qué trabajaste la última vez: decisiones tomadas, archivos modificados, errores encontrados, convenciones aprendidas. Sin nube. Sin telemetría. Todo reside en una base de datos SQLite local en ~/.ctx-memory/store.db.

Cómo funciona

Session starts
      │
 Shell wrapper intercepts 'claude' (or codex/gemini/opencode)
      │
 MCP server starts alongside — injects prior context, exposes memory tools
      │
 During session: hooks fire on every tool call, messages buffered
      │
 Session ends (clean exit / Ctrl-C / crash)
      │
 Layer 1 → Layer 2 → Layer 3 pipeline runs
      │
      ├── Layer 1: extract weighted messages, events, decisions, errors, keywords
      ├── Layer 2: compress to ≤500-token digest, store in DB
      └── Layer 3: merge digest into project memory doc (markdown)

Next session: agent reads prior context automatically

Related MCP server: remem

Instalación

npm install -g ctx-memory
ctx-memory setup

El asistente de configuración:

  1. Detecta qué herramientas están instaladas (claude, codex, gemini, opencode)

  2. Pregunta cuáles integrar

  3. Escribe configuraciones de hook para cada herramienta

  4. Crea enlaces simbólicos de envoltorio en ~/.ctx-memory/bin/

  5. Añade ~/.ctx-memory/bin a tu PATH a través de ~/.bashrc / ~/.zshrc

Reinicia tu shell (o ejecuta source ~/.bashrc), luego usa tus herramientas normalmente: la memoria es automática.

Comandos

ctx-memory setup                        # interactive setup wizard
ctx-memory status                       # show configuration and stats
ctx-memory projects list                # list all projects
ctx-memory projects show <name>         # print full memory doc for a project
ctx-memory projects forget <name>       # reset memory (keeps sessions)
ctx-memory projects forget <name> --hard  # delete all sessions + data

Servidor MCP

El servidor MCP se ejecuta como un proceso secundario junto a cada sesión de herramienta y expone seis herramientas:

Herramienta

Descripción

store_message

Almacena un mensaje de conversación para procesarlo al finalizar la sesión

store_event

Registra un evento de llamada a herramienta (edición de archivo, comando bash, etc.)

search_context

Busca en sesiones pasadas por similitud semántica o palabras clave

get_project_memory

Devuelve el documento completo de memoria del proyecto

list_sessions

Lista las sesiones recientes con objetivos y resultados

end_session

Finaliza una sesión y ejecuta la tubería de Capa 1 → Capa 2 → Capa 3

Para usar el servidor MCP de forma independiente con Claude Code, añade a ~/.claude/settings.json:

{
  "mcpServers": {
    "ctx-memory": {
      "command": "node",
      "args": ["/path/to/ctx-memory/dist/src/mcp/index.js"]
    }
  }
}

Arquitectura

src/
  cli/        — setup wizard, status, projects commands
  db/         — SQLite schema + CRUD (projects, sessions, events, digests, memory)
  layer1/     — pure message/event processing → weighted Layer1Output
  layer2/     — digest compression to ≤500 tokens → Layer2Digest
  layer3/     — merge digest into ProjectMemory markdown
  mcp/        — MCP server (handlers + stdio entry point)
  wrapper/    — shell wrapper (intercepts tool invocation, manages session lifecycle)
  hooks/      — hook config writers for Claude / Gemini / OpenCode

Restricciones clave:

  • Todas las funciones de Capa 1/2/3 son puras: sin E/S, nunca lanzan errores

  • Los resúmenes se mantienen por debajo de 500 tokens

  • La búsqueda vectorial utiliza embeddings all-MiniLM-L6-v2 a través de sqlite-vec (384 dimensiones)

  • Base de datos SQLite única, modo WAL, claves foráneas activadas

Desarrollo

npm run build       # tsc → dist/
npm run dev         # tsc --watch
npm run test:run    # vitest (single-shot, 476+ tests)
npm test            # vitest (watch mode)

La ruta de la base de datos es por defecto ~/.ctx-memory/store.db. Puedes sobrescribirla con CTX_MEMORY_DB_PATH.

Licencia

MIT

Available Tools

6 tools
end_sessionC

Signal session end — runs Layer 1→2→3 pipeline, stores digest, updates project memory.

Inputs: outcome (required), project_id (required), session_id (required), exit_code.

get_project_memoryC

Return the full project memory document (architecture, conventions, recent work, known issues).

Inputs: project_id.

list_sessionsC

List recent sessions for a project with their goals and outcomes.

Inputs: limit, project_id.

search_contextB

Search past sessions for context relevant to a query. Uses embedding similarity, falls back to keyword search.

Inputs: query (required), limit, project_id.

store_eventC

Record a tool call event (file edit, bash command, etc.) for the current session.

Inputs: tool (required), session_id (required), args, result, +1 more.

store_messageC

Buffer a conversation message for session-end processing.

Inputs: role (required), index (required), content (required), session_id (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

    TDQS

    B3.3/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: session lifecycle (end_session), memory retrieval (get_project_memory, list_sessions, search_context), and data recording (store_event, store_message). No overlap in functionality.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern using snake_case, e.g., end_session, get_project_memory, store_event. The naming is predictable and uniform.

    Tool Count5/5

    With 6 tools, the server covers core memory management operations without bloat. The count is well-scoped for the domain of session context and project memory.

    Completeness5/5

    The tool surface covers the full typical lifecycle: recording events/messages, searching and retrieving context, listing sessions, and finalizing sessions with memory updates. No obvious gaps.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers