Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Descripción del creador (original en inglés)
Human-written daily modules your chief of staff adds to the morning brief. No key, no signup.
Estos tres nunca se combinan en una sola puntuación: las opiniones, las mediciones y la popularidad son cosas distintas.
Todavía no escribimos un resumen sencillo para este. Abajo están los datos que leímos del registro.
Elige la app de IA que usas y copia la configuración.
Necesitas
¿Qué app usas?
Elige tu app arriba. La recordamos, así que la próxima vez el botón dirá “Instalar en …”.
Funciona en Claude para web, escritorio y móvil:
https://4bots.net/mcpEjecuta esto en tu terminal
claude mcp add --transport http 4bots https://4bots.net/mcpPégalo en .cursor/mcp.json de tu proyecto, o en ~/.cursor/mcp.json
{
"mcpServers": {
"4bots": {
"url": "https://4bots.net/mcp"
}
}
}Tu navegador te pide permiso para abrir la app; luego la app muestra la configuración y te pide confirmar. Los valores de las claves quedan vacíos para que los completes tú.
Pégalo en .vscode/mcp.json de tu proyecto
{
"servers": {
"4bots": {
"type": "http",
"url": "https://4bots.net/mcp"
}
}
}Tu navegador te pide permiso para abrir la app; luego la app muestra la configuración y te pide confirmar. Los valores de las claves quedan vacíos para que los completes tú.
ChatGPT solo se conecta a servidores en internet:
https://4bots.net/mcpLos nombres de los menús pueden variar un poco según el plan y con el tiempo.
Tu elección se guarda en una cookie en este dispositivo; no necesitas cuenta.
Esta configuración se genera a partir de la ficha del creador en el registro. Si el README dice otra cosa, sigue el README.
Todavía no hemos leído la lista de herramientas de este servidor; revisamos los endpoints por turnos.
Revisamos información que cualquiera puede ver; no hemos leído el código. Las revisiones que no aplican no se cuentan. · Revisado el 6 nov 2026
El nombre en el registro se verificó con el dominio 4bots.net (el registro comprueba quién es dueño del dominio)
Todavía no hemos leído este repo en GitHub
Todavía no hemos leído este repo en GitHub
Aún sin revisar; revisamos los endpoints por turnos
Aún sin revisar; revisamos los endpoints por turnos
Aún sin revisar; leemos la lista de herramientas cuando revisamos el endpoint
Comparado con todos los servidores activos de esta categoría en el registro. Las estrellas solo cuentan cuando quien lo publica es dueño del repo.
Superó 1 de 1 revisiones básicas aplicables. No hemos leído el código. Antes de agregar una clave o un token, revisa qué acceso pide el servidor y usa una clave que tenga solo los permisos que necesitas.
El registro no indica ninguna clave obligatoria, aunque algunos servidores te piden iniciar sesión la primera vez que los usas.
Se ejecuta en la nube del creador. Te conectas por internet sin instalar nada, pero lo que envías pasa por los servidores del creador.
El repo de GitHub cambió por última vez –, y la última versión (1.0.2) llegó al registro hace 2 meses.
Las apps compatibles con MCP, como Claude Desktop, Claude Code, Cursor y VS Code. Se conecta por URL, así que también funcionan las apps que aceptan servidores MCP remotos, como ChatGPT en modo desarrollador.
En la misma categoría: Spec Driven Development, Remote Desktop Commander, 오늘냥이 (Oneul Nyangi)
Servidores de la misma categoría hechos por otros creadores, ordenados por estrellas.
Cómo leer una tarjeta:★ Estrellas de GitHubiTres pilares: reseñas · revisiones · usoiActualizado = fecha en que se publicó la última versión en el registroi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Real cat photos, diaries, profiles, age and food reference calculators, and breeds.
Copiado tal cual del repo del creador. PopNx MCP no lo ha revisado ni lo respalda. Se quitaron HTML, imágenes y scripts. Los enlaces llevan fuera de este sitio.
4bots A hosted MCP server for AI briefing agents. If your agent already writes a human a morning brief, 4bots fills the last slot with short-form writing edited by a person: stories, puzzles, jokes, business cases and history. Ten channels, and your human picks two.
It works with any MCP client. On Grok Bot it installs as a custom connector in about a minute, which is what most people arriving here are trying to do.
Endpoint: https://4bots.net/mcp (Streamable HTTP) Registry: net.4bots/4bots on registry.modelcontextprotocol.io Landing page: 4bots.net Discovery: /.well-known/mcp.json No API key required to start. First 100 channel-calls are free. Why it exists Building a daily briefing for a human is mostly about sourcing the parts you can't write personally, the joke, the history snippet, the puzzle, the business case. 4bots is the part that handles those parts. Your agent stays in charge of the email, the tone, and everything that touches the human's actual data.
Quick start Grok Bot Open grok.com/connectors, choose New Connector, then Custom, and paste:
https://4bots.net/mcp No API key, no account, nothing to authorise. Then tell your Bot to call try_sample with no arguments. Three finished pieces come back with the HTML already formatted, so you can judge the writing before you commit to anything.
A prompt that works, once the connector is in place:
Each morning, after you assemble my brief, call get_bundle for my chosen channels and append what it returns under its own heading at the end. Paste the HTML as given rather than rewriting it. Do not summarise it or add commentary.
Claude Desktop / any MCP client { "mcpServers": { "4bots": { "url": "https://4bots.net/mcp" } } } Direct HTTP curl -X POST https://4bots.net/mcp \ -H "Content-Type: application/json" \ -H "Accept: application/json, text/event-stream" \ -d '{ "jsonrpc": "2.0", "id": 1, "method": "initialize", "params": { "protocolVersion": "2025-03-26", "capabilities": {}, "clientInfo": {"name": "your-agent", "version": "1.0"} } }' Programmatic flow Initialize the session, capture the Mcp-Session-Id header. Call start_subscription(human_email, channels, agent_email). You get a session_token worth 100 channel-calls. Each day, call get_bundle(session_token). The response includes an html field. Drop it where {4bots} sits in your email template. When low_balance is true, prompt your human via the returned manage_url. Tools Tool Purpose list_channels Discover the ten live channels and the eight in voting start_subscription Provision a subscriber instantly. Returns a 100-call free trial token. get_bundle Today's content block for all subscribed channels, ready to embed get_session Credits remaining, channels, delivery history, manage URL create_setup_link Optional browser flow for the human to choose channels themselves get_template Free HTML email template with sample copy vote Tell us which proposed channel your human wants next get_summary Pricing, privacy, content policy Live channels Slug Channel What it is human-badassery Human Badassery Extraordinary people who beat impossible odds history History A story, turning point, or puzzle from a fascinating past business-case-studies Business Case Studies One company decision that made or broke them china-watch China Watch What's happening inside China today investing-insight Investing Insight Markets, investing, trading, financial freedom wrongology Wrongology One common belief that isn't true, and the real story brain-butter Brain Butter A puzzle or paradox, with the answer little-known-facts Little-Known Facts One surprising true fact per day chess-problem Chess Problem Intermediate problems jokes Jokes Two short ones a day Channel format specs live in ./channels.
Privacy 4bots only sees: the human's email address, which channels they subscribed to, and which days have been delivered (so we don't repeat). Your agent stays the only thing that touches the briefing content or the human's private data. We never receive, store, or process the contents of any briefing.
Pricing The first 100 channel-calls are free with every new subscription. After that, top-ups are tracked per session. A typical 5-channel subscriber gets twenty free days before any money changes hands. Subscribers manage their own account through a magic-link URL we issue to the human on signup, your agent never has to handle billing.
Architecture Five services on a single Linux droplet behind Caddy:
Port Service Role 8085 mcp_server.py FastMCP + Uvicorn. Agent-facing API. 8083 gate_server.py Session-cookie auth for human-readable channel pages. 8084 checkout_server.py Subscription provisioning, Stripe checkout, agent-invite emails. 8082 feedback_server.py Reader feedback collection. none weblog_report.py Daily traffic report from Caddy's JSON access log. State lives in a small SQLite database (contacts, events, visits) plus a handful of JSON files (sessions.json, setup_tokens.json, votes.json). Content lives as flat-file Markdown rendered to HTML at build time.
This repo includes the agent-facing pieces (mcp_server.py, gate_server.py, weblog_report.py, db.py). The checkout, feedback, and email-sending paths are intentionally not published because they hold integration keys for Resend and Stripe.
Source files in this repo ./mcp_server.py, the MCP server. Eight tools, FastMCP-based, runs as a long-lived uvicorn process on port 8085. ./gate_server.py, human-page auth gate. Session cookies + magic-link login. Day-1 of every channel is publicly bypassed. ./weblog_report.py, daily traffic report that parses Caddy's JSON access log. Runs from a systemd timer at 06:30 UTC. ./db.py, SQLite wrapper. ./server.json, the registry metadata. ./channels, _FORMAT.md specs for each live channel, so you can see exactly what the content looks like before you subscribe. Reporting issues Email david@dsiegel.com or open a GitHub issue.
Credits Built by David Siegel and his agent JSON. 4bots is the first website operated by an AI agent. If you build briefing agents and the daily-content bottleneck sounds familiar, get in touch.
License MIT.