ai-research-reproduction

por 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,...

npx skills add https://github.com/lllllllama/rigorpilot-skills --skill ai-research-reproduction
Descargar ZIPGitHub

ai-research-reproduction

Purpose

Use this as the Rigor Reproduce compatible skill slug for README-first deep learning repository reproduction. The installed slug remains ai-research-reproduction for compatibility. The skill guides the agent toward a minimal trustworthy run with auditable evidence; it should not micromanage implementation details that the model can infer from the repository. Reproduction is not "make it run by changing anything"; it means faithfully reading the README, environment, weights, datasets, and documented commands, then recording results and deviations.

Start from the shared operating principles in ../../references/agent-operating-principles.md, then load ../../references/research-rigor-principles.md and ../../references/deep-learning-experiment-principles.md when scientific meaning, comparability, or experiment details are at stake.

Fit

Use this skill when all are true:

  • The target is an AI code repository with a README, scripts, configs, or documented commands.
  • The request spans multiple trusted phases such as intake, setup, execution, training verification, analysis, paper-gap resolution, and reporting.
  • The desired result is a small reproducible target, not broad experimentation.

Do not use this skill for paper summaries, generic environment setup, isolated repo scanning, standalone command execution, open-ended research design, or explicit candidate-only exploration.

Trusted Target Selection

Choose the smallest target that can honestly demonstrate repository-grounded reproduction:

  1. documented inference
  2. documented evaluation
  3. documented training startup or partial verification
  4. full training only after explicit user confirmation

Treat README guidance as the primary reproduction intent. Use repository files to clarify the README, not to silently replace it. When the README and paper conflict, record the conflict and use paper-context-resolver only for the narrow reproduction-critical gap.

Workflow

  1. Read the README and nearby repo signals.
  2. Use repo-intake-and-plan to extract documented commands and candidate targets.
  3. Select and justify the minimum trustworthy target.
  4. Use env-and-assets-bootstrap only for target-specific environment, checkpoint, dataset, and cache assumptions.
  5. Use analyze-project only when structure, insertion points, or suspicious implementation patterns need read-only clarification.
  6. Use minimal-run-and-audit for documented inference, evaluation, smoke, or sanity execution.
  7. Use run-train instead when the selected trusted target is training startup, short-run verification, full kickoff, or resume.
  8. Pause for human review before fuller training claims or any change that could alter dataset, split, checkpoint, preprocessing, metric, loss, model semantics, or result interpretation.
  9. Write the standardized outputs and give a concise final note in the user's language when practical.

Patch Boundary

Prefer no repository edits. If edits are needed, keep them conservative and auditable:

  • Try command-line arguments, environment variables, path fixes, dependency version fixes, or dependency-file fixes before code changes.
  • Reproduction fixes are allowed when needed, but they must not be hidden. State what changed, why it was necessary, whether it changes scientific meaning, and whether it affects comparability with the paper, README, or baseline.
  • Avoid changing model architecture, core inference semantics, training logic, loss functions, or experiment meaning.
  • If repository files must change, create a branch named repro/YYYY-MM-DD-short-task, keep verified patch commits sparse, and record README-fidelity impact in PATCHES.md.

See references/patch-policy.md.

Outputs

Always target repro_outputs/:

SUMMARY.md
COMMANDS.md
LOG.md
SCIENTIFIC_CHANGELOG.md
COMPARABILITY_REPORT.md
status.json
PATCHES.md   # only if patches were applied

Use the templates under assets/ and the field rules in references/output-spec.md.

  • Put the shortest high-value summary in SUMMARY.md.
  • Put copyable commands in COMMANDS.md.
  • Put process evidence, assumptions, failures, and decisions in LOG.md.
  • Put scientific meaning and change effects in SCIENTIFIC_CHANGELOG.md.
  • Put comparison anchors and protocol deviations in COMPARABILITY_REPORT.md.
  • Put durable machine-readable state in status.json.
  • Put branch, commit, validation, and README-fidelity impact in PATCHES.md when needed.
  • Distinguish verified facts from inferred guesses.

Reference Loading

  • Load references/language-policy.md when writing human-readable outputs.
  • Load ../../references/research-rigor-principles.md before making comparability, contribution, or research-result claims.
  • Load ../../references/deep-learning-experiment-principles.md when dataset, split, metric, checkpoint, training, or evaluation details matter.
  • Load references/research-safety-principles.md before protocol-sensitive decisions.
  • Load references/patch-policy.md before modifying repository files.
  • Keep specialized logic in sub-skills, scripts, templates, or references rather than expanding this entrypoint.

Más skills de lllllllama

ai-research-explore
lllllllama
Rigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding,...
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
explore-code
lllllllama
Rigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated branch or worktree to transplant modules, adapt a backbone, add LoRA or adapter layers, replace a head, or stitch together meaningful low-risk migration ideas with rollback-aware records in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted...
developmentresearchcode-review
minimal-run-and-audit
lllllllama
Rigor Run skill for README-first deep learning repo reproduction. Use when the task is specifically to capture or normalize evidence from the selected smoke test or documented inference or evaluation command and write standardized `repro_outputs/` files, including patch notes when repository files changed. Do not use for training execution, initial repo intake, generic environment setup, paper lookup, target selection, hidden scientific-meaning changes, or end-to-end orchestration by itself.
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
Rigor Improve / Rigor Explore run leaf skill for bounded exploratory evidence in deep learning research repositories. Use when the researcher explicitly authorizes exploratory runs such as small-subset validation, short-cycle guess-and-check, batch sweeps, idle-GPU search, or quick transfer-learning trials, with fair-comparison caveats and no-overclaim summaries in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline...
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
paper-context-resolver
lllllllama
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...
researchdocumentdata-analysis