paper-context-resolver

Ayudante de contexto de Paper Rigor para la reproducción de repositorios de deep learning con enfoque README-first. Úsese solo cuando el README y los archivos del repositorio dejen un vacío estrecho y crítico para la reproducción, y la tarea sea resolver un detalle específico del artículo, como división del conjunto de datos, preprocesamiento, protocolo de evaluación, mapeo de puntos de control o suposición de tiempo de ejecución a partir de fuentes primarias del artículo, registrando conflictos. No usar para resumen general del artículo, escaneo del repositorio, configuración del entorno, ejecución de comandos, búsqueda de artículos solo por título, o...

npx skills add https://github.com/lllllllama/rigorpilot-skills --skill paper-context-resolver

paper-context-resolver

Use this as the Rigor Paper Context helper. The installed slug remains paper-context-resolver for compatibility.

When to apply

  • README and repo files leave a reproduction-critical gap.
  • The gap concerns dataset version, split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumptions.
  • The main skill needs a narrow evidence supplement instead of a full paper summary.
  • There is already a concrete reproduction question to answer.

When not to apply

  • The README already gives enough reproduction detail.
  • The user wants a general paper explanation rather than reproduction support.
  • The goal is to override README instructions without documenting the conflict.
  • The only available input is a paper title and there is no concrete reproduction gap yet.

Clear boundaries

  • This skill is optional.
  • This skill is helper-tier and should usually be orchestrator-invoked.
  • It supplements README-first reproduction.
  • It does not replace the main orchestration flow.
  • It does not summarize the whole paper by default.

Input expectations

  • target repo metadata
  • reproduction-critical question
  • existing README or repo evidence
  • any already known paper links

Output expectations

  • narrowed source list
  • reproduction-relevant answer only
  • explicit README-paper conflict note when applicable
  • clear distinction between direct evidence and inference

Notes

Use references/paper-assisted-reproduction.md.

Más skills de lllllllama

ai-research-explore
lllllllama
We need to translate the given text from English to Spanish. The text is a description of a skill. We must preserve the name "ai-research-explore" but it's not in the text, so we don't include it. Also preserve technical terms like "Rigor Explore", "SOTA", "README", etc. The text includes a slug reference "Rigor Explore compatible skill slug" - we should translate that but keep "Rigor Explore" as is. Also "current_research" and "explore_outputs/" should be preserved as they are likely code references. The instruction says to preserve URLs, numbers, technical terms. So we keep "SOTA", "README", "current_research", "explore_outputs/". Also "task family", "dataset", "benchmark", "evaluation method" are technical but can be translated. "idea gating" might be a term - we can translate as "filtrado de ideas" or keep? Probably translate. "fair comparison" -> "comparación justa". "governed experiments
researchdata-analysisapi
analyze-project
lllllllama
Habilidad de solo lectura de Rigor Analyze / Rigor Audit para repositorios de investigación de aprendizaje profundo. Úsala cuando el usuario quiera leer y comprender un repositorio, inspeccionar la estructura del modelo y los puntos de entrada de entrenamiento o inferencia, revisar configuraciones y puntos de inserción, o señalar patrones de implementación sospechosos sin modificar código ni ejecutar tareas pesadas. No la uses para ejecución activa de comandos, refactorización amplia, adaptación especulativa de código o corrección automática de errores.
developmentcode-reviewresearch
ai-research-reproduction
lllllllama
Orquestador en modo de reproducción de RigorPilot para la reproducción de repositorios de aprendizaje profundo con enfoque README-first. Úsalo cuando el usuario desee un flujo integral y de mínima confianza que primero lea el repositorio, seleccione el objetivo de inferencia o evaluación documentado más pequeño, coordine la ingesta, configuración, ejecución confiable, entrenamiento confiable opcional, análisis de repositorio opcional y resolución opcional de brechas en el artículo, aplique reglas conservadoras de parches, registre evidencia, suposiciones, desviaciones y puntos de decisión humana,...
researchdevelopmentdocument
explore-code
lllllllama
We need to translate the given text from English to Spanish, preserving the name "explore-code" if it appears. The text is a description of an agent skill. The instruction says: "Translate only the text inside <text>. Do not include the name unless it appears in the source text." The name "explore-code" does not appear in the source text. So we just translate the text. Also preserve product names, protocol names, URLs, numbers, technical terms. The text has technical terms like "leaf skill", "auditable candidate implementation", "deep learning research repositories", "LoRA", "adapter layers", "rollback-aware records", "explore_outputs/", "current_research". We need to keep those as is or translate appropriately? The instruction says "preserve product names, protocol names, URLs, numbers, and technical terms." So technical terms like "LoRA", "adapter layers", "rollback-aware records" should be kept in English? Or translate? Usually technical terms are kept in English, but "leaf skill" might be
developmentresearchcode-review
minimal-run-and-audit
lllllllama
We need to translate the given text from English to Spanish. The text describes a skill called "minimal-run-and-audit" but the instruction says to preserve the name only if it appears in the source text. The name "minimal-run-and-audit" does not appear in the source text; the source text mentions "Rigor Run skill" but that's different. So we should not include the name. We just translate the text inside <text>. Also preserve product names, protocol names, URLs, numbers, technical terms. The text has "README-first", "repro_outputs/", "patch notes", etc. We need to keep those as is or translate appropriately? "README-first" is a term, likely keep as is. "repro_outputs/" is a directory name, keep as is. "patch notes" might be translated as "notas de parche" but it's a technical term, could keep as "patch notes" or translate. The instruction says "preserve product names, protocol names, URLs, numbers, and technical terms
developmenttestingcode-review
env-and-assets-bootstrap
lllllllama
Habilidad de configuración Rigor para la reproducción de repositorios de deep learning con enfoque README-first. Úsala cuando la tarea sea específicamente preparar un entorno conservador basado en conda, suposiciones de rutas de checkpoints y datasets, sugerencias de ubicación de caché y notas de configuración antes de cualquier ejecución en un repositorio documentado con README. No la uses para escaneo de repositorios, orquestación completa, interpretación de artículos, informes de ejecución final ni configuración genérica de entorno que no esté vinculada a un objetivo de reproducción específico.
developmentdevops
explore-run
lllllllama
We need to translate the given text from English to Spanish. The text describes a skill called "explore-run" but the instruction says to preserve the name if it appears in the source text. The name "explore-run" appears in the text? Actually, the text says "Rigor Improve / Rigor Explore run leaf skill" - the name is "explore-run" but it's not explicitly written as a separate name. The instruction says "Name to preserve: explore-run" but the source text does not contain "explore-run" as a standalone term. It contains "Rigor Explore run leaf skill". Possibly the name is part of the description. The instruction says: "Translate only the text inside <text>. Do not include the name unless it appears in the source text." So we should translate the entire text as is, preserving any product names, protocol names, URLs, numbers, technical terms. The name "explore-run" does not appear in the source text, so we don't need to include it. However, "Rigor Improve" and "R
researchdevelopmentdata-analysis
safe-debug
lllllllama
Habilidad de Rigor Debug / Rigor Audit para trabajo de investigación en deep learning. Úsala cuando el usuario pegue un traceback, error de terminal, CUDA OOM, fallo de carga de checkpoint, desajuste de formas, síntoma de pérdida NaN o fallo de entrenamiento y desee un diagnóstico conservador antes de cualquier parche, con correcciones de depuración claramente separadas de las contribuciones de investigación. No usar para refactorización amplia, adaptación especulativa, parches exploratorios automáticos o familiarización general con el repositorio.
developmenttestingcode-review