| name | carrera |
| version | 1.0.0 |
| description | Pack de carrera (es): proceso de seleccion, entrevistas, negociacion, CV, onboarding, red. Topics: acta-formal, cierre-conversacion, cv-cover-optimizer, cv-enhancement, follow-up-email, gratitud-post-proceso, negociacion-oferta, onboarding-90-dias, proceso-seleccion-orchestrator, red-y-referencias, simulador-entrevista, validar-liquidacion-co. |
| params | {"topic":{"enum":["acta-formal","cierre-conversacion","cv-cover-optimizer","cv-enhancement","follow-up-email","gratitud-post-proceso","negociacion-oferta","onboarding-90-dias","proceso-seleccion-orchestrator","red-y-referencias","simulador-entrevista","validar-liquidacion-co"],"required":true,"infer":"from user request; ask only if ambiguous"},"depth":{"enum":["quick","deep"],"default":"quick"}} |
| routes | {"acta-formal":"references/acta-formal.md","cierre-conversacion":"references/cierre-conversacion.md","cv-cover-optimizer":"references/cv-cover-optimizer.md","cv-enhancement":"references/cv-enhancement.md","follow-up-email":"references/follow-up-email.md","gratitud-post-proceso":"references/gratitud-post-proceso.md","negociacion-oferta":"references/negociacion-oferta.md","onboarding-90-dias":"references/onboarding-90-dias.md","proceso-seleccion-orchestrator":"references/proceso-seleccion-orchestrator.md","red-y-referencias":"references/red-y-referencias.md","simulador-entrevista":"references/simulador-entrevista.md","validar-liquidacion-co":"references/validar-liquidacion-co.md"} |
carrera
Router skill for the Spanish career cluster. Resolve topic, then Read EXACTLY
ONE playbook from routes: and follow it. Never load the whole cluster. [DOC]
When to use
Trigger on candidate-side career work: selection-process tracking, mock
interviews, offer evaluation, CV/cover optimization, follow-ups, 30/60/90
onboarding, reference networks, gratitude notes, formal acta, conversation
closeout, or Colombian liquidación validation. [DOC] If the ask is hiring-side
(JD authoring, sourcing, screening rubrics for an employer), this is out of
scope — route to a recruiting skill instead. [INFERENCIA]
Inputs → Outputs
- In: request + supplied evidence (emails, notes, CV, offer, payslip);
topic (required, inferred); depth (default quick). [CONFIG]
- Out: exactly the chosen playbook's contract, each non-obvious claim tagged
from one tag family:
[DOC] [CONFIG] [INFERENCIA] [SUPUESTO]. [DOC]
Routing
- Match the request to one topic enum. Ambiguous between two → ask one
disambiguating question; never guess silently. [INFERENCIA]
- Read the single
routes: playbook for that topic. Do not preload siblings. [DOC]
depth=deep → apply exhaustively, verify at each step; quick →
essentials only, same gates. [CONFIG]
Spine (every playbook): Discover → Analyze → Execute → Validate, under
constitution v6.0.0 enforcement, evidence tags, and the script-first rule. [CONFIG]
Validation gate (before returning)
Anti-patterns
- Loading several playbooks "to compare" — defeats the router. [INFERENCIA]
- Fabricating salary/market figures or settlement amounts to fill gaps. [INFERENCIA]
- Mixing tag families,
{WEB} without citation, or free-handing past a
playbook's script-first steps. [DOC]
Assets
Router-level support lives in assets/ (see assets/README.md):
assets/quality-rubric.json (guardian gates) and assets/routing-keywords.json
(topic disambiguation). Per-topic policy assets stay in each playbook. [CONFIG]