Ctrl K
ComunidadGuías…
GITHUB · ANALIZADO

rails-informant

Self-hosted error monitoring for Rails with MCP server for agentic workflows

Ficha actualizada 23 sept 2026, 10:37 · upstream 23 sept 2026, 10:04 · release v0.5.3

Ver códigoProblemas y ayudaRepositorio ↗
00

Qué es rails-informant

Descripción editorial construida a partir del repositorio, su documentación y sus señales de mantenimiento.

Rails Informant is a self-hosted error monitoring solution built specifically for AI agents and developer tools. It captures exceptions, stores them in your application's database with rich context including backtraces, breadcrumbs, request data, and user information. The platform exposes an MCP (Model Context Protocol) server that allows AI agents to query, triage, and resolve production errors directly within their workflow without needing a traditional dashboard or external tooling. The architecture is designed for agent-native workflows where the agent itself serves as the interface to error resolution.

CASOS DE USO
  • AI agents can list, inspect, and resolve production errors directly from the MCP server
  • Self-hosted monitoring with complete data sovereignty — no data leaves your infrastructure
  • Eliminate dashboard dependency by making the agent itself the primary interaction point
  • Context-rich error resolution where every occurrence includes full stack traces, request context, and user information
  • Auto-resolve stale errors on deployment via deploy notifications and verification checks
  • Selective capture of specific exception types without affecting normal Rails HTTP 4xx responses
PUNTOS DESTACADOS
  • Agent-native interface — the MCP server enables AI agents to list, inspect, resolve, and fix errors directly in their workflow context
  • Self-hosted with zero external dependencies — all data stays within your infrastructure
  • Automatic noise suppression for common Rails exceptions while preserving real bugs like job failures or race conditions
  • Contextual fingerprinting that preserves sensitive information (email, IP) only when explicitly opted-in via configuration
  • Auto-resolve stale errors on deployment with automatic re-opening of resolved issues upon regression
  • Developer-friendly CLI and MCP tools for manual investigation alongside agent automation
PRECAUCIONES
  • No built-in rate limiting — must be implemented externally using Rack::Attack or similar libraries
  • Only captures exceptions raised by Rails itself, not errors from third-party gems or external dependencies
  • Requires a database to store error occurrences (Redis is optional but recommended for performance)
  • The MCP server requires HTTPS and authentication tokens in production environments
  • No built-in retry logic — job failures must be handled manually via the CLI or agent workflow
01

Perfil de infraestructura

Estimación inicial derivada del repositorio y de los servicios detectados.

CPU1 vCPUrecomendación inicial
MEMORIA1 GBRAM estimada
DISCO20 GBalmacenamiento base
COMPLEJIDADlownivel de operación
02

Despliegue detectado

Qué señales reales de instalación hemos encontrado.

Despliegue manual

Install RailsInformant by adding it to your Gemfile, running bundle install, then executing bin/rails generate rails_informant:install and bin/rails db:migrate. Configure authentication tokens in credentials for API access and MCP server communication. Set environment variables INFORMANT_API_TOKEN, INFORMANT_PRODUCTION_URL, and INFORMANT_PRODUCTION_TOKEN for the MCP server to authenticate requests. Enable noise suppression by default but customize ignored exceptions per application needs. For multi-environment setups, configure separate URLs and tokens per environment. The gem includes a doctor command that reports whether installed files are current or stale relative to your local version.

—Dockerfile
—Compose
—One-click
03

Stack y dependencias

Componentes inferidos a partir de manifests, configuración y documentación.

FRAMEWORKS
Rails
SERVICIOS
Redis
BASES DE DATOS
Redis
PUERTOS
No detectados
05

Puede sustituir a

Relaciones detectadas por el análisis del repositorio y su categoría.

06

Experiencia real de la comunidad

Ayuda a contrastar la estimación automática con instalaciones reales.

Todavía no hay datos comunitarios para este proyecto.