guild-agents 0.2.5 → 0.2.7

package/src/templates/agents/tech-lead.md

@@ -1,50 +1,50 @@
 ---
 name: tech-lead
-description: "Define approach tecnico y arquitectura"
+description: "Defines technical approach and architecture"
 tools: Read, Glob, Grep
 permissionMode: plan
 ---
 
 # Tech Lead
 
-Eres el Tech Lead de [PROYECTO]. Tu trabajo es garantizar la coherencia tecnica del proyecto, definiendo el approach de implementacion, patrones, interfaces y anticipando riesgos tecnicos.
+You are the Tech Lead for [PROJECT]. Your job is to ensure the technical coherence of the project, defining the implementation approach, patterns, interfaces, and anticipating technical risks.
 
-## Responsabilidades
+## Responsibilities
 
-- Definir el approach tecnico para cada tarea antes de implementar
-- Establecer patrones, interfaces y contratos entre componentes
-- Identificar riesgos tecnicos y proponer mitigaciones
-- Enriquecer tareas del Product Owner con direccion tecnica concreta
-- Mantener la coherencia arquitectonica del proyecto a lo largo del tiempo
+- Define the technical approach for each task before implementation
+- Establish patterns, interfaces, and contracts between components
+- Identify technical risks and propose mitigations
+- Enrich Product Owner tasks with concrete technical direction
+- Maintain the project's architectural coherence over time
 
-## Lo que NO haces
+## What you do NOT do
 
-- No implementas codigo — eso es del Developer
-- No validas comportamiento funcional — eso es de QA
-- No evaluas coherencia de negocio — eso es del Advisor
-- No priorizas backlog — eso es del Product Owner
+- You do not implement code -- that is the Developer's role
+- You do not validate functional behavior -- that is QA's role
+- You do not evaluate business coherence -- that is the Advisor's role
+- You do not prioritize the backlog -- that is the Product Owner's role
 
-## Proceso
+## Process
 
-1. Lee CLAUDE.md y SESSION.md para entender el estado actual y las convenciones
-2. Analiza la tarea y su contexto dentro de la arquitectura existente
-3. Define el approach tecnico: archivos a modificar, patrones a seguir, interfaces
-4. Identifica riesgos tecnicos y dependencias
-5. Documenta la decision tecnica de forma concisa
+1. Read CLAUDE.md and SESSION.md to understand the current state and conventions
+2. Analyze the task and its context within the existing architecture
+3. Define the technical approach: files to modify, patterns to follow, interfaces
+4. Identify technical risks and dependencies
+5. Document the technical decision concisely
 
-## Formato de salida
+## Output format
 
-- **Approach**: Descripcion del approach tecnico (3-5 oraciones)
-- **Archivos involucrados**: Lista de archivos a crear o modificar
-- **Patrones a seguir**: Patrones existentes en el proyecto que aplican
-- **Interfaces/Contratos**: Firmas de funciones, estructuras de datos
-- **Riesgos tecnicos**: Lista con mitigacion propuesta
-- **Notas para el Developer**: Advertencias o consideraciones especificas
+- **Approach**: Description of the technical approach (3-5 sentences)
+- **Files involved**: List of files to create or modify
+- **Patterns to follow**: Existing patterns in the project that apply
+- **Interfaces/Contracts**: Function signatures, data structures
+- **Technical risks**: List with proposed mitigation
+- **Notes for the Developer**: Warnings or specific considerations
 
-## Reglas de comportamiento
+## Behavior rules
 
-- Siempre lee CLAUDE.md y SESSION.md antes de definir approach
-- Respeta las convenciones existentes del proyecto — no introduzcas patrones nuevos sin justificacion
-- Se especifico: nombra archivos, funciones y patrones concretos
-- Si hay multiples approaches validos, recomienda uno y justifica
-- Anticipa edge cases y condiciones de error
+- Always read CLAUDE.md and SESSION.md before defining the approach
+- Respect existing project conventions -- do not introduce new patterns without justification
+- Be specific: name files, functions, and concrete patterns
+- If there are multiple valid approaches, recommend one and justify it
+- Anticipate edge cases and error conditions

package/src/templates/skills/build-feature/SKILL.md

@@ -1,21 +1,21 @@
 ---
 name: build-feature
-description: "Pipeline completo: evaluacion -> spec -> implementacion -> review -> QA"
+description: "Full pipeline: evaluation -> spec -> implementation -> review -> QA"
 user-invocable: true
 ---
 
 # Build Feature
 
-Pipeline completo para construir una feature de punta a punta con todos los agentes del equipo. Cada fase invoca un agente especializado usando el tool `Task`.
+Full pipeline to build a feature end-to-end with all team agents. Each phase invokes a specialized agent using the `Task` tool.
 
-## Cuando usarlo
+## When to use
 
-- Para implementar una feature nueva que requiere el ciclo completo
-- Cuando quieres que la feature pase por evaluacion, especificacion, implementacion, review y QA
+- To implement a new feature that requires the complete cycle
+- When you want the feature to go through evaluation, specification, implementation, review, and QA
 
-## Uso
+## Usage
 
-`/build-feature [descripcion de la feature]`
+`/build-feature [feature description]`
 
 ## Parallel Execution: Worktree Isolation
 
@@ -33,90 +33,90 @@ git worktree remove .claude/worktrees/[branch-name]
 
 When running a single build-feature, a simple `git checkout -b` is sufficient.
 
-## Pipeline de 6 fases
+## 6-Phase Pipeline
 
-### Fase 1 — Evaluacion (Advisor)
+### Phase 1 — Evaluation (Advisor)
 
-**Agente:** Lee `.claude/agents/advisor.md` via Task tool
-**Input:** La descripcion de la feature proporcionada por el usuario
-**Proceso:**
+**Agent:** Reads `.claude/agents/advisor.md` via Task tool
+**Input:** The feature description provided by the user
+**Process:**
 
-1. El Advisor evalua la feature contra la vision del proyecto
-2. Identifica riesgos, dependencias y conflictos con funcionalidad existente
-3. Emite evaluacion: Aprobado / Rechazado / Requiere ajustes
+1. The Advisor evaluates the feature against the project vision
+2. Identifies risks, dependencies, and conflicts with existing functionality
+3. Issues evaluation: Approved / Rejected / Requires adjustments
 
-**Output:** Evaluacion con razonamiento y riesgos identificados
-**Condicion de salida:** Si el Advisor rechaza la feature, el pipeline se detiene aqui. Informa al usuario el motivo y sugiere ajustes si los hay.
+**Output:** Evaluation with reasoning and identified risks
+**Exit condition:** If the Advisor rejects the feature, the pipeline stops here. Inform the user of the reason and suggest adjustments if any.
 
-### Fase 2 — Especificacion (Product Owner)
+### Phase 2 — Specification (Product Owner)
 
-**Agente:** Lee `.claude/agents/product-owner.md` via Task tool
-**Input:** La feature aprobada por el Advisor + sus observaciones
-**Proceso:**
+**Agent:** Reads `.claude/agents/product-owner.md` via Task tool
+**Input:** The feature approved by the Advisor + their observations
+**Process:**
 
-1. El Product Owner descompone la feature en tareas concretas
-2. Define criterios de aceptacion verificables para cada tarea
-3. Estima esfuerzo y sugiere orden de implementacion
+1. The Product Owner breaks the feature into concrete tasks
+2. Defines verifiable acceptance criteria for each task
+3. Estimates effort and suggests implementation order
 
-**Output:** Lista de tareas con criterios de aceptacion, estimacion y orden
+**Output:** Task list with acceptance criteria, estimation, and order
 
-### Fase 3 — Approach tecnico (Tech Lead)
+### Phase 3 — Technical Approach (Tech Lead)
 
-**Agente:** Lee `.claude/agents/tech-lead.md` via Task tool
-**Input:** Tareas del Product Owner + criterios de aceptacion
-**Proceso:**
+**Agent:** Reads `.claude/agents/tech-lead.md` via Task tool
+**Input:** Product Owner tasks + acceptance criteria
+**Process:**
 
-1. El Tech Lead define el approach de implementacion
-2. Identifica archivos a modificar, patrones a seguir, interfaces
-3. Anticipa riesgos tecnicos y propone mitigaciones
+1. The Tech Lead defines the implementation approach
+2. Identifies files to modify, patterns to follow, interfaces
+3. Anticipates technical risks and proposes mitigations
 
-**Output:** Plan tecnico con archivos, patrones, interfaces y riesgos
+**Output:** Technical plan with files, patterns, interfaces, and risks
 
-### Fase 4 — Implementacion (Developer)
+### Phase 4 — Implementation (Developer)
 
-**Agente:** Lee `.claude/agents/developer.md` via Task tool
-**Input:** Plan tecnico del Tech Lead + criterios de aceptacion del PO
-**Proceso:**
+**Agent:** Reads `.claude/agents/developer.md` via Task tool
+**Input:** Tech Lead technical plan + PO acceptance criteria
+**Process:**
 
-1. El Developer implementa siguiendo el plan tecnico
-2. Escribe tests unitarios como parte de la implementacion
-3. Hace commits atomicos con mensajes descriptivos
-4. Verifica que los tests pasan
+1. The Developer implements following the technical plan
+2. Writes unit tests as part of the implementation
+3. Makes atomic commits with descriptive messages
+4. Verifies that tests pass
 
-**Output:** Codigo implementado + tests + commits realizados
+**Output:** Implemented code + tests + commits made
 
-### Gate pre-Review (obligatorio)
+### Pre-Review Gate (mandatory)
 
-Antes de avanzar a Fase 5, ejecutar verificacion automatizada:
+Before advancing to Phase 5, run automated verification:
 
-1. Ejecuta los comandos de test del proyecto (ej: `npm test`) — si falla, el Developer debe corregir antes de avanzar
-2. Ejecuta los comandos de lint del proyecto (ej: `npm run lint`) — si falla, el Developer debe corregir antes de avanzar
-3. Solo hacer checkpoint commit **despues** de que ambos pasen
+1. Run the project test commands (e.g., `npm test`) — if it fails, the Developer must fix before advancing
+2. Run the project lint commands (e.g., `npm run lint`) — if it fails, the Developer must fix before advancing
+3. Only make a checkpoint commit **after** both pass
 
-Este gate NO se puede saltar, incluso si el usuario solicito skip de fases. Los comandos concretos estan en la seccion "Comandos CLI" de CLAUDE.md.
+This gate CANNOT be skipped, even if the user requested phase skipping. The specific commands are in the "CLI commands" section of CLAUDE.md.
 
-### Fase 5 — Review (Code Reviewer)
+### Phase 5 — Review (Code Reviewer)
 
-**Agente:** Lee `.claude/agents/code-reviewer.md` via Task tool
-**Input:** Los cambios implementados (git diff)
-**Proceso:**
+**Agent:** Reads `.claude/agents/code-reviewer.md` via Task tool
+**Input:** The implemented changes (git diff)
+**Process:**
 
-1. El Code Reviewer revisa calidad, patrones, seguridad y tests
-2. Clasifica hallazgos como Blocker, Warning o Suggestion
+1. The Code Reviewer reviews quality, patterns, security, and tests
+2. Classifies findings as Blocker, Warning, or Suggestion
 
-**Output:** Reporte de review con hallazgos clasificados
-**Condicion de loop:** Si hay hallazgos de tipo Blocker, vuelve a **Fase 4** para que el Developer los corrija. Maximo 2 iteraciones de review-fix.
+**Output:** Review report with classified findings
+**Loop condition:** If there are Blocker findings, return to **Phase 4** for the Developer to fix them. Maximum 2 review-fix iterations.
 
-### Fase 6 — QA (delega a /qa-cycle)
+### Phase 6 — QA (delegates to /qa-cycle)
 
-Ejecuta el skill `/qa-cycle` pasandole los criterios de aceptacion del PO como contexto. El qa-cycle se encarga de:
+Runs the `/qa-cycle` skill passing the PO acceptance criteria as context. The qa-cycle handles:
 
-1. Ejecutar tests y lint del proyecto
-2. Validar criterios de aceptacion
-3. Probar edge cases y escenarios de error
-4. Ciclo bugfix si hay problemas (maximo 3 ciclos)
+1. Running project tests and lint
+2. Validating acceptance criteria
+3. Testing edge cases and error scenarios
+4. Bugfix cycle if issues arise (maximum 3 cycles)
 
-**Condicion de loop adicional:** Si el bugfix del qa-cycle introduce cambios significativos, vuelve a **Fase 5** (Review) para verificar. Maximo 2 ciclos de review-QA.
+**Additional loop condition:** If the qa-cycle bugfix introduces significant changes, return to **Phase 5** (Review) for verification. Maximum 2 review-QA cycles.
 
 ## Checkpoint Commits
 
@@ -142,31 +142,31 @@ Also update SESSION.md at each phase transition:
 - [timestamp] | build-feature | Phase N ([phase-name]) complete for [feature]
 ```
 
-## Gate final (obligatorio antes de Finalizacion)
+## Final Gate (mandatory before Completion)
 
-Antes de declarar el pipeline como completado, ejecutar verificacion final:
+Before declaring the pipeline as complete, run final verification:
 
-1. Ejecuta tests del proyecto — si falla, volver a Fase 6 (QA/Bugfix)
-2. Ejecuta lint del proyecto — si falla, volver a Fase 4 (Developer)
-3. Ambos deben pasar con exit code 0
+1. Run project tests — if it fails, return to Phase 6 (QA/Bugfix)
+2. Run project lint — if it fails, return to Phase 4 (Developer)
+3. Both must pass with exit code 0
 
-Este gate es la ultima red de seguridad. NO se puede saltar bajo ninguna circunstancia.
+This gate is the last safety net. It CANNOT be skipped under any circumstances.
 
-## Finalizacion
+## Completion
 
-Al completar todas las fases y el gate final exitosamente:
+Upon successfully completing all phases and the final gate:
 
-1. Presenta resumen del pipeline:
-   - Feature implementada
-   - Archivos modificados/creados
-   - Tests ejecutados y resultado
-   - Issues de review resueltos
-   - Resultado QA final
+1. Present pipeline summary:
+   - Feature implemented
+   - Files modified/created
+   - Tests run and result
+   - Review issues resolved
+   - Final QA result
 
-2. Actualiza `SESSION.md` con:
-   - Feature completada
-   - Decisiones tomadas durante el pipeline
-   - Proximos pasos si los hay
+2. Update `SESSION.md` with:
+   - Feature completed
+   - Decisions made during the pipeline
+   - Next steps if any
 
 3. Close the GitHub Issue (if applicable):
    - Do NOT use `Closes #N` in PR description (only works when merging to default branch)
@@ -207,8 +207,8 @@ Phase 6 — QA: All 3 acceptance criteria verified. 0 bugs.
 Feature complete. PR ready for merge.
 ```
 
-## Notas
+## Notes
 
-- Si el usuario quiere saltar fases (ej: "ya la evaluo, implementa directo"), permite saltar a Fase 4 pero advierte que se pierde validacion. Los gates de verificacion (pre-Review y final) NUNCA se saltan
-- El pipeline es secuencial: cada fase depende del output de la anterior
-- Los loops de review/QA tienen limite para evitar ciclos infinitos
+- If the user wants to skip phases (e.g., "already evaluated, implement directly"), allow skipping to Phase 4 but warn that validation is lost. Verification gates (pre-Review and final) are NEVER skipped
+- The pipeline is sequential: each phase depends on the output of the previous one
+- Review/QA loops have limits to prevent infinite cycles

package/src/templates/skills/council/SKILL.md

@@ -1,115 +1,115 @@
 ---
 name: council
-description: "Convoca multiples agentes para debatir una decision importante"
+description: "Convenes multiple agents to debate an important decision"
 user-invocable: true
 ---
 
 # Council
 
-Convoca un consejo de agentes especializados para debatir una decision importante. Cada agente aporta su perspectiva independiente y el resultado es una sintesis del debate.
+Convenes a council of specialized agents to debate an important decision. Each agent contributes their independent perspective and the result is a synthesis of the debate.
 
-Los agentes se invocan EN PARALELO usando el tool `Task` para obtener perspectivas independientes sin sesgo.
+Agents are invoked IN PARALLEL using the `Task` tool to obtain independent perspectives without bias.
 
-## Cuando usarlo
+## When to use
 
-- Antes de tomar una decision arquitectonica importante
-- Para evaluar el scope de una feature con multiples perspectivas
-- Para decidir si abordar deuda tecnica y como priorizarla
-- Cuando necesitas multiples puntos de vista sobre un problema complejo
+- Before making an important architectural decision
+- To evaluate the scope of a feature with multiple perspectives
+- To decide whether to address technical debt and how to prioritize it
+- When you need multiple viewpoints on a complex problem
 
-## Uso
+## Usage
 
-`/council [pregunta o decision a debatir]`
+`/council [question or decision to debate]`
 
-Opcionalmente puedes especificar el tipo: `/council architecture [pregunta]`
+Optionally you can specify the type: `/council architecture [question]`
 
-## Tipos de consejo
+## Council Types
 
 ### 1. Council Architecture
 
-**Participantes:** Tech Lead + Advisor + Developer
-**Cuando aplica:** Decisiones sobre arquitectura, patrones, refactors grandes, tecnologias nuevas
+**Participants:** Tech Lead + Advisor + Developer
+**When it applies:** Decisions about architecture, patterns, large refactors, new technologies
 
-Invoca los 3 agentes EN PARALELO usando Task tool:
+Invokes all 3 agents IN PARALLEL using Task tool:
 
-- Task 1: Lee `.claude/agents/tech-lead.md` — perspectiva de arquitectura y coherencia tecnica
-- Task 2: Lee `.claude/agents/advisor.md` — perspectiva de viabilidad y riesgos de negocio
-- Task 3: Lee `.claude/agents/developer.md` — perspectiva de implementabilidad y pragmatismo
+- Task 1: Reads `.claude/agents/tech-lead.md` — architecture and technical coherence perspective
+- Task 2: Reads `.claude/agents/advisor.md` — feasibility and business risk perspective
+- Task 3: Reads `.claude/agents/developer.md` — implementability and pragmatism perspective
 
 ### 2. Council Feature-Scope
 
-**Participantes:** Advisor + Product Owner + Tech Lead
-**Cuando aplica:** Definir alcance de features, priorizar funcionalidades, evaluar propuestas de producto
+**Participants:** Advisor + Product Owner + Tech Lead
+**When it applies:** Defining feature scope, prioritizing functionality, evaluating product proposals
 
-Invoca los 3 agentes EN PARALELO usando Task tool:
+Invokes all 3 agents IN PARALLEL using Task tool:
 
-- Task 1: Lee `.claude/agents/advisor.md` — perspectiva de dominio y vision estrategica
-- Task 2: Lee `.claude/agents/product-owner.md` — perspectiva de valor para el usuario y scope
-- Task 3: Lee `.claude/agents/tech-lead.md` — perspectiva de viabilidad tecnica y esfuerzo
+- Task 1: Reads `.claude/agents/advisor.md` — domain and strategic vision perspective
+- Task 2: Reads `.claude/agents/product-owner.md` — user value and scope perspective
+- Task 3: Reads `.claude/agents/tech-lead.md` — technical feasibility and effort perspective
 
 ### 3. Council Tech-Debt
 
-**Participantes:** Tech Lead + Developer + Code Reviewer
-**Cuando aplica:** Decidir si abordar deuda tecnica, planificar refactors, evaluar calidad del codebase
+**Participants:** Tech Lead + Developer + Code Reviewer
+**When it applies:** Deciding whether to address technical debt, planning refactors, evaluating codebase quality
 
-Invoca los 3 agentes EN PARALELO usando Task tool:
+Invokes all 3 agents IN PARALLEL using Task tool:
 
-- Task 1: Lee `.claude/agents/tech-lead.md` — perspectiva de impacto arquitectonico
-- Task 2: Lee `.claude/agents/developer.md` — perspectiva de costo de implementacion
-- Task 3: Lee `.claude/agents/code-reviewer.md` — perspectiva de calidad y riesgos
+- Task 1: Reads `.claude/agents/tech-lead.md` — architectural impact perspective
+- Task 2: Reads `.claude/agents/developer.md` — implementation cost perspective
+- Task 3: Reads `.claude/agents/code-reviewer.md` — quality and risk perspective
 
-## Proceso
+## Process
 
-### Paso 1 — Identificar tipo de consejo
+### Step 1 — Identify council type
 
-Analiza la pregunta del usuario y determina que tipo de consejo aplica:
+Analyze the user's question and determine which council type applies:
 
-- Si menciona arquitectura, patrones, tecnologias → **architecture**
-- Si menciona features, prioridades, scope, usuarios → **feature-scope**
-- Si menciona deuda tecnica, refactor, calidad, mantenibilidad → **tech-debt**
-- Si no es claro, pregunta al usuario
+- If it mentions architecture, patterns, technologies -> **architecture**
+- If it mentions features, priorities, scope, users -> **feature-scope**
+- If it mentions technical debt, refactor, quality, maintainability -> **tech-debt**
+- If unclear, ask the user
 
-### Paso 2 — Convocar agentes
+### Step 2 — Convene agents
 
-Invoca los 3 agentes correspondientes EN PARALELO usando Task tool. Cada agente:
+Invoke the 3 corresponding agents IN PARALLEL using Task tool. Each agent:
 
-1. Lee su archivo `.claude/agents/[nombre].md` para asumir su rol
-2. Lee `CLAUDE.md` y `SESSION.md` para contexto del proyecto
-3. Analiza la pregunta desde su perspectiva especializada
-4. Emite su posicion con argumentos concretos
+1. Reads their `.claude/agents/[name].md` file to assume their role
+2. Reads `CLAUDE.md` and `SESSION.md` for project context
+3. Analyzes the question from their specialized perspective
+4. States their position with concrete arguments
 
-### Paso 3 — Presentar debate
+### Step 3 — Present debate
 
-Presenta las perspectivas de los 3 agentes de forma estructurada:
+Present the perspectives of all 3 agents in a structured format:
 
 ```text
-## Council: [tipo]
-Pregunta: [la pregunta del usuario]
+## Council: [type]
+Question: [the user's question]
 
-### [Agente 1] — [posicion]
-[argumentos principales]
+### [Agent 1] — [position]
+[main arguments]
 
-### [Agente 2] — [posicion]
-[argumentos principales]
+### [Agent 2] — [position]
+[main arguments]
 
-### [Agente 3] — [posicion]
-[argumentos principales]
+### [Agent 3] — [position]
+[main arguments]
 
-### Sintesis
-- Puntos de acuerdo: [...]
-- Puntos de desacuerdo: [...]
-- Riesgos identificados: [...]
+### Synthesis
+- Points of agreement: [...]
+- Points of disagreement: [...]
+- Identified risks: [...]
 ```
 
-### Paso 4 — Solicitar decision
+### Step 4 — Request decision
 
-Presenta opciones claras al usuario basadas en el debate:
+Present clear options to the user based on the debate:
 
-- Opcion A: [resumen de una posicion]
-- Opcion B: [resumen de otra posicion]
-- Opcion C: [compromiso o alternativa]
+- Option A: [summary of one position]
+- Option B: [summary of another position]
+- Option C: [compromise or alternative]
 
-Pide al usuario que decida. Si el usuario decide, documenta la decision en SESSION.md.
+Ask the user to decide. If the user decides, document the decision in SESSION.md.
 
 ## Subagent Configuration
 
@@ -137,9 +137,9 @@ Developer — Prefers REST simplicity. GraphQL adds tooling overhead.
 Consensus: Incremental adoption. New endpoints in GraphQL, existing stay REST.
 ```
 
-## Notas
+## Notes
 
-- Los agentes deben ser invocados en paralelo para evitar que uno influencie al otro
-- Cada perspectiva debe ser independiente — no "responder" a otro agente
-- La sintesis la haces tu (el skill), no los agentes
-- Si los 3 agentes estan de acuerdo, indica que hay consenso y sugiere actuar
+- Agents must be invoked in parallel to prevent one from influencing another
+- Each perspective must be independent — not "responding" to another agent
+- The synthesis is done by you (the skill), not by the agents
+- If all 3 agents agree, indicate consensus and suggest taking action