Ek-Chuah MCP

Ek-Chuah MCP

Read-only MCP server that serves the Archivo Epistemico (epistemic graph), enabling AI to read and query the epistemic path autonomously via tools like search, get_via, resolve, and get_necesidad.

Category
Visit Server

README

Ek-Chuah MCP -- MCP-AEC (lector C3)

Servidor MCP read-only que sirve el graph_aec (Archivo Epistemico) en la nube: cualquier IA lee la via epistemica de forma autonoma, y por I3 cada lectura trae su porque. Hermano del substrato Ek-Chuah (misma topologia que concept-sediment / concept-sediment-mcp: substrato y servidor MCP son repos separados).

Contrato de origen: concept-sediment/docs/ek_chuah/ -- CONVERGENCIA (proceso 2), LECTOR_C3 (esta spec), SCHEMA_YAML_AEC, y REPORTE_CodeCS_a_CodeMCP_paso4_habilitado.

Arquitectura (camino B -- Guardian 2026-06-27)

La nube no recibe la proyeccion ya construida; recibe el log (forma Q) y reconstruye la proyeccion aqui. Honra forma Q / I1, evita el desync de proyeccion one-way, y reusa el stack concept-sediment-mcp.

  Scripts/AEC/log/inscripciones.jsonl   (durable local, la VERDAD -- CodeCS)
        |  exportador del log  (SOL JOINT a CodeCS; bytes nivel-1 NO cruzan)
        v
  aec_log (JSONB append-only)           copia replicada del log en la nube
        |  projection_build.rebuild_projection  (idempotente, det.)
        v
  7 tablas graph_aec + afirmacion.embedding   proyeccion REGENERABLE
        |  queries.py (read-only)
        v
  4 tools MCP  ->  cualquier IA

Tools (4 core, todas read-only)

Tool Job Entrada
aec_search "?ya investigue X?" -- indice anti-re-investigacion (embeddings + ILIKE) query
aec_get_via una afirmacion + su via ascendente completa (I3) af_id
aec_resolve un referente -> cadena de versiones (I4) url | referente_id | content_hash
aec_get_necesidad traza descendente: que trajo una indagacion y que sobrevivio nec_id

aec_search devuelve cada hit con stub de via (nunca dato pelon); aec_get_via expande la via completa. 5a tool aec_divergencia diferida (LECTOR_C3 §8).

Membrana (el candado que no se rompe)

  • No escribe. Read-only por construccion + get_readonly_session (SET TRANSACTION READ ONLY) + rol de BD solo-lectura en Railway. Aportar va por Estratega, fuera de aqui.
  • No baja URLs. Sin web; la url es metadato forense que devuelve, no que fetcha.
  • No sirve nivel 1. Devuelve content_hash, nunca los bytes del snapshot.

Estructura

server.py            FastMCP + Starlette + 4 tools aec_*
queries.py           lectura read-only (ports de proyeccion.py)
projection_build.py  rebuild de aec_log -> 7 tablas + embeddings (camino B)
schema.sql           DDL: aec_log + 7 tablas proyeccion + pgvector
db.py                engine; get_readonly_session (defensa membrana)
normaliza_url.py     espejo vendorizado de D-ver-1 (canonical URL)

Local

cp .env.example .env          # DATABASE_URL (Postgres+pgvector) + GEMINI_API_KEY
python projection_build.py --replicate ../AEC/log/inscripciones.jsonl  # carga aec_log
python projection_build.py --rebuild                                   # reconstruye + embeddings
python server.py                                                       # lector en :8000

Falsador I1 (rebuild = identico, excluye embedding): python projection_build.py --dump debe coincidir con Ek-Chuah/proyeccion.dump_logico sobre el mismo log.

Deploy (gateado por Guardian)

El deploy del MCP-AEC invalida sesiones MCP (CONVERGENCIA §9). Antes: YAMLs de cierre reviewed + MTV + baselines si cambian tools. Railway apunta a este repo, no al substrato. Post-deploy: reiniciar cliente -> reconectar MCP.

Pendiente JOINT

El exportador del log local -> aec_log es jurisdiccion CodeCS (SOL del borde, docs/ek_chuah/SOL_EXPORTADOR_LOG_AEC_2026-06-27.md). Mientras tanto, projection_build.py --replicate lo simula desde el archivo local para pruebas.

Recommended Servers

playwright-mcp

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.

Official
Featured
TypeScript
Magic Component Platform (MCP)

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.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

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.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

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.

Official
Featured
TypeScript
Kagi MCP Server

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.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

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.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured