1. Conduid
  2. Science
  3. Pipeline
MCP server · Science

Pipeline

Stateful Pipeline framework for MCP servers — type-safe state management, declarative tool chaining, token-efficient design

Unclaimed science
34Low

Scored 8 days ago · breakdown

About Pipeline

Pipeline is an MCP server in the Science category: stateful Pipeline framework for MCP servers — type-safe state management, declarative tool chaining, token-efficient design.

Install

From source
git clone https://github.com/ThomasPraun/mcp-pipeline

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Pipeline

Powered by Claude · Grounded in docs

I know everything about Pipeline. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

mcp-pipeline

Skill orquestador para Claude Code que guía paso a paso la creación de MCP servers de alta calidad para cualquier API o servicio.

Qué hace

Coordina 15+ skills a través de 7 fases — desde la investigación del API hasta la publicación del package — siguiendo el patrón "coordinate, never execute": el pipeline dirige, los skills especializados ejecutan.

Phase 1: Discovery       → Investigar API, mapear tools, elegir lenguaje
Phase 2: Architecture    → Diseñar schemas, estructura, plan de implementación
Phase 3: Scaffolding     → Inicializar proyecto, dependencias, server base
Phase 4: Implementation  → TDD por tool, code review, verificación
Phase 5: Testing         → MCP Inspector, transport, auth
Phase 6: Evaluation      → 10 QA pairs + harness automático (accuracy ≥ 70%)
Phase 7: Distribution    → README, packaging, publicación

Instalación

npx skills add ThomasPraun/mcp-pipeline@mcp-pipeline -g -y

Uso

En Claude Code:

> mcp pipeline
> build mcp
> create mcp server
> mcp status
> next phase

Características

  • Soporte TypeScript y Python — selección automática basada en el SDK del API target y el stack del usuario
  • Orquesta mcp-builder — inyecta las guías de MCP correctas en cada fase
  • State management — tracking de progreso con .claude/mcp-project/status.json
  • Dashboard visual — progreso por fase con porcentaje de completitud
  • Agent Teams — 4 ventanas de paralelismo para acelerar el desarrollo
  • Navegación no-lineal — skip ahead, go back, resume desde skill parcial
  • Auto-install — detecta skills faltantes y ofrece instalación

Dependencias

Requiere el skill mcp-builder instalado. El pipeline verifica automáticamente y ofrece instalar skills faltantes.

Estructura

mcp-pipeline/
├── SKILL.md                              # Orquestador principal
└── references/
    ├── phase-1-discovery.md              # API research + language decision
    ├── phase-2-architecture.md           # Tool schemas + project structure
    ├── phase-3-scaffolding.md            # Project setup + dependencies
    ├── phase-4-implementation.md         # TDD + code review
    ├── phase-5-testing.md                # MCP Inspector + transport
    ├── phase-6-evaluation.md             # QA pairs + evaluation harness
    ├── phase-7-distribution.md           # README + packaging
    ├── install-commands.md               # Skill install commands
    ├── skill-injection-protocol.md       # Subagent delegation protocol
    └── language-decision-matrix.md       # TypeScript vs Python decision tree

Licencia

MIT

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About Pipeline

How do I install Pipeline?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Pipeline safe to use with an AI agent?

Its trust score is 34 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Pipeline still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.