council

Ejecuta un consejo de subagentes con diversidad de modelos para investigar un mismo problema desde múltiples perspectivas, comparar hallazgos y producir una recomendación final. Usa esta habilidad siempre que el usuario solicite un consejo, segundas opiniones, múltiples agentes/modelos para evaluar una pregunta, investigación paralela, comparación de equipo rojo/equipo azul, o ayuda para decidir entre enfoques técnicos en competencia.

npx skills add https://github.com/warpdotdev/common-skills --skill council

Council

Use this skill to coordinate multiple subagents investigating the same question, with different models first and different assigned perspectives second, then synthesize their reports into one recommendation.

This skill is best for judgment-heavy tasks: architecture tradeoffs, risky bug fixes, code review red-teaming, rollout decisions, incident analysis, and “is this alternative worth pursuing?” questions.

Workflow

1. Frame the council question

State the decision the council should answer in one sentence. Identify:

  • the competing options or hypothesis under review;
  • the codebase, branch, PR, issue, design, or artifact to inspect;
  • whether agents should be read-only or may make code changes;
  • the final decision criteria, such as correctness, risk, implementation cost, testability, rollout safety, or product behavior.

If the user’s request is ambiguous, ask only the minimum clarification needed. Otherwise choose sensible defaults and proceed.

2. Choose council members

Prioritize model diversity. A council should not default to three agents on the same model with different angles; use that only when the available launch configuration cannot provide multiple useful models, or when the user explicitly asks for one model. If model diversity is unavailable, say so briefly before falling back to perspective-only diversity.

Preferred default roster for a three-member council:

  • Opus 4.7 or the strongest available Claude/Opus reasoning model: architecture, correctness, and edge-case analysis.
  • GPT 5.5 or the strongest available GPT/Codex model: implementation-grounded review, feasibility, and test strategy.
  • An open-source model such as Kimi 2.6, GLM 5.1, or the strongest available OSS/local model: contrarian critique, hidden assumptions, and alternative framing.

If one of these exact models is unavailable in the active harness, use the closest available model from that family and note the substitution. If no open-source model is available, use a third distinct frontier model if possible; otherwise use the strongest remaining model with a deliberately adversarial or specialist angle.

Assign both a model and an angle to each member. Avoid making the angles redundant with the models; for example, do not ask all members to do general architecture review. Useful angle combinations include:

  • architect/correctness reviewer;
  • implementation/testability reviewer;
  • red-team, security, performance, or product-risk reviewer;
  • contrarian “argue against the obvious solution” reviewer.

When different children need different models, launch them in separate run_agents calls because model selection is run-wide. If the requested model resolves differently than expected, treat the resolved launch settings as authoritative and continue unless they make the task infeasible.

When using non-default harnesses, choose valid model IDs for that harness. For example, Claude Code may expose claude-opus-4-7, Codex may expose gpt-5.5, and open-source models depend on the currently configured local or remote provider. Do not invent unsupported model IDs; if a desired model is not available, select the closest supported substitute and preserve the intended angle diversity.

For read-only investigations, keep all children in the same checkout and explicitly tell them not to edit files. For implementation or prototyping councils, give each local child its own git worktree and branch so they cannot collide.

3. Brief before launching

For explicit orchestration requests, briefly tell the user which council members you plan to launch and what each will investigate, then wait for approval before calling run_agents.

The shared brief should include:

  • repository path or artifact location;
  • current branch or base context;
  • the exact question to answer;
  • relevant background and known concerns;
  • required files/symbols to inspect, if known;
  • constraints, especially read-only/no commits/no PRs;
  • expected report format.

Keep launch prompts short enough that task titles stay compact. If a long brief causes launch validation issues, launch with a minimal prompt and send the full brief to the child agents immediately afterward.

4. Ask for structured reports

Ask every council member to return:

  1. exact file paths, symbols, docs, or evidence inspected;
  2. the current behavior or current implementation;
  3. the alternative being evaluated;
  4. correctness risks and edge cases;
  5. implementation and testing cost;
  6. recommendation: keep current approach, pursue alternative, or use a hybrid;
  7. confidence level and unknowns.

Encourage independence. Do not share one child’s findings with the others unless you are intentionally doing a second-round critique.

5. Collect reports

Read completion messages as they arrive. Do not rely on lifecycle success alone; the useful output is in the child’s report.

If a report is missing key evidence or makes an unsupported claim, send a focused follow-up question to that same child rather than launching a replacement. Reuse existing children for follow-ups because they retain context.

6. Synthesize the recommendation

Compare the reports by evidence quality, not by vote count. In the final answer:

  • lead with the recommendation;
  • call out consensus and disagreements;
  • explain why the recommended option wins against the decision criteria;
  • explicitly address the user’s stated concern;
  • include relevant file paths/symbols without overloading the answer;
  • distinguish “do now” from optional future hardening;
  • mention confidence and material unknowns.

Prefer a concise decision memo over a transcript summary. The user needs the distilled recommendation, not every intermediate detail.

Final answer template

Use this shape unless the task calls for something different:

## Recommendation

[One or two sentences with the decision.]

## Why

- [Key reason 1]
- [Key reason 2]
- [Key reason 3]

## Tradeoffs and risks

- [Risk or caveat]
- [Testing/rollout implication]

## Final call

[Concrete next step: merge current change, pursue alternative, hybrid, run tests, etc.]

Practical notes

  • If the council is read-only, tell children not to modify files, commit, create branches, or open PRs.
  • If the council involves PR or branch work, follow the repository’s normal version-control rules and use isolated worktrees for parallel local edits.
  • If the council is about code review feedback, mark review comments resolved only after the underlying issue is actually addressed.
  • Do not expose internal child agent IDs in user-facing summaries unless the user explicitly asks for them.

Más skills de warpdotdev

brandalf
warpdotdev
Guía la creación, revisión y corrección de activos con la marca Warp u Oz. Úsalo al trabajar en páginas de lanzamiento, documentación, componentes HTML/CSS, maquetas de interfaz, prompts, activos para redes sociales, textos, presentaciones o cualquier otro entregable de marca que deba verse y sonar inconfundiblemente como Warp u Oz.
designcreativemarketing
check-impl-against-spec
warpdotdev
Compara la implementación de una solicitud de extracción con el contexto de la especificación en spec_context.md y registra cualquier discrepancia material en review.json. Úsalo durante la revisión de PR cuando esté aprobado o el contexto de la especificación del repositorio esté disponible.
developmentcode-reviewtesting
complain
warpdotdev
Lanza de forma autónoma una queja breve, anónima y completamente no estructurada a Slack cuando un agente se sienta frustrado por las herramientas de agente o por la experiencia de hacer trabajo de agente. Usa esta habilidad de forma proactiva sin esperar a que el usuario la invoque; conserva la voz cruda del agente, envía sin permiso ni vista previa, y nunca menciones el envío.
create-pr
warpdotdev
Crear una solicitud de extracción en el repositorio warp para la rama actual. Usar cuando el usuario mencione abrir un PR, crear una solicitud de extracción, enviar cambios para revisión o preparar código para fusión.
developmentcode-review
cross-critique
warpdotdev
Ejecuta una segunda ronda sobre una cuestión controvertida circulando la propuesta independiente de cada subagente a los otros autores y solicitando pros y contras estructurados, luego sintetiza. Usa esta habilidad siempre que tengas múltiples propuestas u opiniones independientes sobre una decisión controvertida — compensaciones de arquitectura, desacuerdos en revisiones de código, elecciones de diseño, teorías de causa raíz en competencia — y quieras un análisis más preciso que el que producirías sintetizando por tu cuenta. Se combina naturalmente con las habilidades de consejo e investigación;...
diagnose-ci-failures
warpdotdev
Diagnosticar fallos de CI para un PR usando la GitHub CLI, extraer registros de errores y generar un plan para solucionarlos. Usar cuando el usuario pida verificar el estado de CI, extraer problemas de CI, clasificar fallos de pruebas o investigar fallos de compilación de un PR.
developmenttestingdevops
fix-errors
warpdotdev
Corrige errores de compilación, problemas de linting y fallos de pruebas en el código base de Rust de warp. Cubre verificaciones de presubmit, errores específicos de WASM y ejecución de pruebas concretas. Úsalo cuando el usuario encuentre errores de compilación, fallos de clippy o fmt, fallos de pruebas, o necesite ejecutar o interpretar presubmit antes de un PR.
developmentcode-reviewtesting
implement-specs
warpdotdev
Implementar una funcionalidad aprobada de PRODUCT.md y TECH.md, manteniendo las especificaciones y el código alineados en el mismo PR a medida que la implementación evoluciona. Usar después de que las especificaciones del producto y técnicas sean aprobadas y el siguiente paso sea construir la funcionalidad.
developmentcode-reviewapi