Automated summary
Automatically generates technical project documentation and a personalized 30/60/90 onboarding plan, then activates a buddy agent that answers the new developer's questions.
Strengths
- Produces a structured documentation set of 10 to 12 files (architecture, glossary, local setup, PR workflow, deployment, postmortems, FAQ)
- Tailors the plan and access checklist to role and seniority (junior, mid, senior), including a dedicated first-PR guide
- The AI buddy is explicitly restricted to guidance and takes no decisions, which bounds its use
- Checks clear prerequisites (project CLAUDE.md, repository access, equipo.md) before starting the analysis
Limitations
- Experimental maturity: generated documents may vary in quality and completeness
- Aborts when the project has no CLAUDE.md or repository access is missing; without equipo.md the output stays partial
- Does not cover HR/GDPR aspects, which are handled by the complementary team-onboarding skill
Use it whenever a developer joins an existing project and you want to cut ramp-up from several weeks down to a few days.
Avoid it on a project with no accessible repository or CLAUDE.md, or when the need is administrative onboarding and compliance rather than technical onboarding.
Security analysis
SafeThe skill describes a documentation-generation and onboarding workflow (scan repo, read files, produce markdown docs, configure a buddy agent). No declared tools, but the implied operations are bounded file reads and writes within projects/{proyecto}/onboarding/. There are no curl|sh, rm -rf, base64 payloads, or secret-sending instructions, and it includes explicit safety guardrails (buddy does not make decisions, only orients). Execution risk is low; the only mild concern is aggregation of potentially sensitive project context into generated docs, which is mitigated by git-ignoring.
- •Skill reads project files, git log, and README/ADRs to auto-generate documentation; no network calls, no shell one-liners, no exfiltration or destructive commands are specified.
- •Generated documentation is placed in a git-ignored directory and explicitly flagged as potentially containing sensitive project info — a responsible handling practice, though it means sensitive context is aggregated into files.
Examples
Ejecuta /onboarding-dev Laura Méndez --rol backend --seniority jr --proyecto pagos-api y genera la documentación del proyecto más su plan 30/60/90 y la guía del primer PR.Onboarding técnico de un nuevo senior en el proyecto checkout-web: quiero el overview de alto nivel, las decisiones técnicas clave y un plan donde su primer PR llegue en el día 1 o 2.Ya están generados los documentos de onboarding de projects/pagos-api/onboarding/. Activa el buddy IA para que responda las dudas de la nueva incorporación usando esa documentación como base.name: onboarding-dev description: "Usar cuando se incorpora un desarrollador nuevo al proyecto y necesita buddy IA." summary: | Onboarding tecnico con Buddy IA: auto-genera documentacion del proyecto, plan personalizado 30/60/90 y agente buddy de 3 capas. Input: nombre + rol. Output: guia + plan + buddy activo. maturity: experimental context: fork agent: tech-writer category: "sdd-framework" tags: ["onboarding", "buddy-ia", "documentation", "ramp-up"] priority: "medium"
Skill: Onboarding Dev — Buddy IA
Seguridad:
@docs/rules/domain/autonomous-safety.md— el buddy IA NO toma decisiones, solo orienta. Basado en: Guía de Onboarding con IA de Manfred (v3.0) — modelo de agente en 3 capas. Complementa:team-onboarding(RGPD, evaluación) — esta skill se enfoca en onboarding técnico.
Cuándo usar esta skill
- Se incorpora un nuevo desarrollador al proyecto
- Se quiere acelerar el ramp-up de 4-8 semanas a 5-10 días
- Se necesita generar documentación del proyecto automáticamente
- Se busca reducir la dependencia del buddy humano para preguntas repetitivas
Qué produce
Fase 1: Auto-análisis (10-12 documentos)
projects/{proyecto}/onboarding/
├── 01-arquitectura-alto-nivel.md ← Diagrama + explicación de servicios y flujos
├── 02-mapa-equipos-ownership.md ← Quién lleva cada área/sistema
├── 03-glosario-interno.md ← Vocabulario, siglas, nombres de proyectos
├── 04-herramientas-accesos.md ← Checklist de configuraciones y accesos
├── 05-setup-local.md ← Guía paso a paso + troubleshooting típico
├── 06-flujos-pr-estandares.md ← Convenciones de commit, PR, review, CI/CD
├── 07-como-se-despliega.md ← Entornos, releases, rollback, permisos
├── 08-decisiones-tecnicas-clave.md ← Por qué se eligió X, qué trade-offs se aceptaron
├── 09-historico-incidencias.md ← Incidencias importantes y postmortems resumidos
├── 10-faq-onboarding.md ← Preguntas más repetidas y sus respuestas
├── 11-calidad-procesos.md ← Linting, formato, patrones, anti-patrones, DoD
└── 12-expectativas-rampup.md ← Qué se espera a los 7 días, 30 días, 90 días
Fase 2: Plan personalizado
projects/{proyecto}/onboarding/
├── {nombre}-plan-30-60-90.md ← Objetivos medibles por periodo
├── {nombre}-checklist-accesos.md ← Checklist personalizada de accesos y configuraciones
└── {nombre}-primer-pr.md ← Guía para el primer PR (paso a paso)
Fase 3: Buddy IA activo
El agente queda disponible como buddy interactivo para responder preguntas del nuevo miembro usando la documentación generada como base de conocimiento.
Prerequisitos
1. Proyecto configurado en projects/{nombre}/CLAUDE.md → si no: ❌ ABORT
2. Acceso al repositorio del proyecto → si no: ❌ ABORT
3. equipo.md actualizado con miembros actuales → si no: ⚠️ continuar con limitaciones
Flujo completo
PM ejecuta /onboarding-dev {nombre} --rol {rol} --seniority {jr|mid|sr} --proyecto {nombre}
↓
Validar prerequisitos
↓
═══ FASE 1: AUTO-ANÁLISIS DEL PROYECTO ═══
↓
Analizar repositorio:
- Leer CLAUDE.md del proyecto
- Escanear estructura de directorios
- Identificar tecnologías (package.json, .csproj, requirements.txt, etc.)
- Leer README, CONTRIBUTING, ARCHITECTURE si existen
- Analizar git log (últimos 3 meses) para entender actividad reciente
- Identificar decisiones técnicas en ADRs si existen
↓
Generar los 10-12 documentos en projects/{proyecto}/onboarding/
↓
Mostrar resumen de documentos generados → PM valida
↓
═══ FASE 2: PLAN PERSONALIZADO ═══
↓
Adaptar según rol y seniority:
- Junior: más detalle, más checkpoints, tareas más pequeñas, más pair programming
- Mid: balance entre autonomía y guía, primer PR en día 2-3
- Senior: high-level overview, primer PR en día 1-2, foco en decisiones técnicas
↓
Generar:
- Plan 30/60/90 con objetivos medibles
- Checklist de accesos personalizada (GitHub, Jira, SonarCloud, VPN, Figma, etc.)
- Guía de primer PR paso a paso
↓
═══ FASE 3: BUDDY IA ACTIVO ═══
↓
Configurar agente buddy con:
- Base de conocimiento: documentos generados en Fase 1
- System prompt con guardarraíles de Manfred (ver abajo)
- Disponible vía /onboarding-ask {pregunta}
Buddy IA — Comportamiento (Manfred v3.0)
El agente buddy responde con explicaciones breves y accionables, cita fuentes internas (archivo + sección), señala nivel de confianza (alto/medio/bajo), y deriva a personas cuando el tema es sensible (seguridad, compliance). No inventa políticas ni detalles técnicos.
3 capas: Orientación (quién/dónde) → Ejecución (setup/PR/deploy) → Contexto (decisiones técnicas pasadas y trade-offs).
Guardarraíles anti-patrón: No usar IA para decidir sin entender. Validar decisiones de arquitectura con el equipo. Si no puedes explicar el código, no lo subas. Preguntar primero a la IA, luego al equipo.
Métricas de ramp-up
| Métrica | Junior | Mid | Senior | |---------|--------|-----|--------| | Tiempo hasta primer PR | ≤ 5 días | ≤ 3 días | ≤ 2 días | | Primer PR aprobado sin cambios | ≤ 10 días | ≤ 5 días | ≤ 3 días | | Primera contribución significativa | ≤ 30 días | ≤ 15 días | ≤ 7 días | | Independencia (tasks sin spec) | ≤ 30 días | ≤ 15 días | ≤ 7 días | | Confianza autoreportada (≥7/10) | Día 15 | Día 10 | Día 5 |
Cuándo NO usar
- El miembro ya lleva >2 semanas productivo (usar
/team-evaluatepara actualizar perfil) - Para perfiles no técnicos (usar
team-onboardinggenérico) - Si no existe CLAUDE.md del proyecto (crear primero con
/project-new)
Almacenamiento
Documentación generada en projects/{proyecto}/onboarding/ — directorio en .gitignore por defecto (puede contener info sensible del proyecto).
Plan personalizado en projects/{proyecto}/onboarding/{nombre}-*.md — igualmente git-ignorado.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/developer-onboarding-with-ai-mrx3xzyz/download' --output '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md.part' && mv '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md.part' '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md' && test -s '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/onboarding-dev' && cp -R '.skill-downloads/developer-onboarding-with-ai-mrx3xzyz' '.claude/skills/onboarding-dev'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/developer-onboarding-with-ai-mrx3xzyz'openskills read 'developer-onboarding-with-ai-mrx3xzyz'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/developer-onboarding-with-ai-mrx3xzyz/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
Supabase Docs Authoring
Documentation
Write, edit, and review Supabase documentation content including guides, tutorials, reference docs, and partials. Follow style guides and MDX/TOML validation rules.
Create Architecture Decision Record
Documentation
Create a new Architecture Decision Record with sequential numbering, register it in the AgentDB graph, and link related ADRs.
Plan update with deep research
Documentation
Updates a plan document by launching parallel subagents to analyze a topic across all compiler passes, then consolidates findings into the plan doc.