AI Context Manager
Manages .ai/ workspace folder, syncs assets with a cloud service, and generates environment-specific bootstrap files for AI context management.
README
AI Context Manager MCP
Servidor MCP (Model Context Protocol) en Python que actúa como sync agent entre el workspace local y la aplicación cloud.
¿Qué hace?
- Gestiona la carpeta
.ai/en tu workspace (skills, prompts, specs, contexto, bootstrap) - Sincroniza assets con la app cloud (
cloud_sync pull/push) - Genera
MODEL_BOOTSTRAP.mdadaptado al entorno (vscode, claude, opencode, cli, generic) - Mantiene
.gitignoreactualizado para no commitear el contexto local
Requisitos
- Python 3.11+
- pip / pipx
Instalación
# Con pipx (recomendado, instala en entorno aislado)
pipx install .
# O con pip en un virtualenv
python -m venv .venv
source .venv/bin/activate # Linux/macOS
.venv\Scripts\activate # Windows
pip install -e .
Variables de entorno (obligatorias para sync cloud)
# Linux/macOS
export AI_CONTEXT_MANAGER_BASE_URL="https://cloud.example.com"
export AI_CONTEXT_MANAGER_TOKEN="pat_xxx"
# Windows (PowerShell)
$env:AI_CONTEXT_MANAGER_BASE_URL="https://cloud.example.com"
$env:AI_CONTEXT_MANAGER_TOKEN="pat_xxx"
# Windows (cmd)
set AI_CONTEXT_MANAGER_BASE_URL=https://cloud.example.com
set AI_CONTEXT_MANAGER_TOKEN=pat_xxx
Arrancar el servidor MCP
# Modo stdio (para clientes MCP como Claude Desktop, OpenCode, etc.)
python -m mcp_server
# O usando el script instalado
ai-context-manager serve
Configuración en VS Code (tasks.json)
Crear .vscode/tasks.json en tu proyecto:
{
"version": "2.0.0",
"tasks": [
{
"label": "AI Context Manager: Start MCP",
"type": "shell",
"command": "python -m mcp_server",
"options": {
"env": {
"AI_CONTEXT_MANAGER_BASE_URL": "https://cloud.example.com",
"AI_CONTEXT_MANAGER_TOKEN": "pat_xxx"
}
},
"problemMatcher": []
},
{
"label": "AI Context Manager: Sync (pull)",
"type": "shell",
"command": "ai-context-manager cloud-sync --direction pull",
"options": {
"env": {
"AI_CONTEXT_MANAGER_BASE_URL": "https://cloud.example.com",
"AI_CONTEXT_MANAGER_TOKEN": "pat_xxx"
}
},
"problemMatcher": []
}
]
}
Configuración en Claude Desktop
Añadir en claude_desktop_config.json:
{
"mcpServers": {
"ai-context-manager": {
"command": "python",
"args": ["-m", "mcp_server"],
"env": {
"AI_CONTEXT_MANAGER_BASE_URL": "https://cloud.example.com",
"AI_CONTEXT_MANAGER_TOKEN": "pat_xxx"
}
}
}
}
Configuración en OpenCode
Añadir en tu config de OpenCode:
{
"mcp": {
"servers": {
"ai-context-manager": {
"command": "python",
"args": ["-m", "mcp_server"],
"env": {
"AI_CONTEXT_MANAGER_BASE_URL": "https://cloud.example.com",
"AI_CONTEXT_MANAGER_TOKEN": "pat_xxx"
}
}
}
}
}
Setup inicial de un proyecto
# 1. Inicializar .ai/ en el workspace
ai-context-manager init --mode workspace
# 2. Vincular con proyecto cloud
ai-context-manager cloud-link --project-key my-project
# 3. Descargar assets del cloud
ai-context-manager cloud-sync --direction pull
# 4. Asegurar .gitignore
ai-context-manager ensure-gitignore
Tools disponibles (MCP)
| Tool | Descripción |
|---|---|
init_storage |
Inicializa .ai/ en workspace o global |
ensure_gitignore |
Añade .ai/ al .gitignore |
scan_repo |
Escanea el repo buscando assets IA |
list_assets |
Lista assets (skills/prompts/specs/context) |
register_asset |
Registra un asset existente en el registry |
move_asset |
Mueve un asset actualizando el registry |
remove_asset |
Elimina un asset del registry |
create_skill |
Crea un nuevo skill desde template |
create_prompt |
Crea un nuevo prompt desde template |
create_spec |
Crea una nueva spec desde template |
generate_bootstrap |
Genera MODEL_BOOTSTRAP.md para el entorno |
cloud_project_link |
Vincula workspace con proyecto cloud |
cloud_sync |
Sincroniza assets (pull: cloud→local, push: local→cloud) |
cloud_pull_backup |
Descarga un backup específico del cloud |
Resources disponibles (MCP)
| Resource | Descripción |
|---|---|
registry:// |
Contenido completo del registry.json |
context://bootstrap |
Contenido del MODEL_BOOTSTRAP.md |
skills://<id> |
Contenido de un skill por ID |
prompts://<id> |
Contenido de un prompt por ID |
specs://<id> |
Contenido de una spec por ID |
Estructura local generada
.ai/
registry.json # fuente de verdad local
context/
AI_GUIDELINES.md
MODEL_BOOTSTRAP.md # generado por generate_bootstrap
skills/
*.md
prompts/
*.md
specs/
*.md
templates/
skill.md
prompt.md
spec.md
.sync/
state.json # estado de sync (hashes/ETags)
project.json # binding local_path <-> project_key
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.