repo-intake-and-plan

作者: lllllllama

用于README优先的深度学习仓库复现的严格输入助手。当任务明确为扫描仓库、读取README及常见项目文件、提取文档化命令、分类推理、评估和训练候选方案,并向主协调器返回最小可信复现计划时使用。不用于环境设置、资源下载、命令执行、最终报告、论文查阅或端到端协调。

npx skills add https://github.com/lllllllama/rigorpilot-skills --skill repo-intake-and-plan

repo-intake-and-plan

Use this as the Rigor Intake helper. The installed slug remains repo-intake-and-plan for compatibility.

When to apply

  • At the beginning of README-first reproduction work.
  • When the main skill needs a fast map of repo structure and documented commands.
  • When inference, evaluation, and training candidates must be classified conservatively.
  • When the user explicitly wants to inspect the repo first and not run anything yet.

When not to apply

  • When execution has already started and the task is now about running commands or writing outputs.
  • When the target is not a repository-backed reproduction task.
  • When the user only wants paper interpretation without repo inspection.
  • When the user already has a selected documented command and only needs setup or execution.

Clear boundaries

  • This skill scans and plans.
  • This skill is helper-tier and should usually be orchestrator-invoked.
  • It does not install environments.
  • It does not prepare large assets.
  • It does not execute substantive reproduction commands.
  • It does not decide high-risk patching.

Input expectations

  • Target repository path.
  • Access to README and common project files if present.
  • Optional user hints about desired priority, such as inference-first.

Output expectations

  • concise repo structure summary
  • documented command inventory
  • inferred candidate categories: inference, evaluation, training, other
  • minimum trustworthy reproduction recommendation
  • notable ambiguity or risk list

Notes

Use references/repo-scan-rules.md and helper scripts under scripts/.

来自 lllllllama 的更多技能

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
针对深度学习研究仓库的Rigor Analyze / Rigor Audit只读技能。当用户希望阅读和理解仓库、检查模型结构与训练或推理入口点、审查配置与插入点,或在无需修改代码或运行繁重任务的情况下标记可疑实现模式时使用。不适用于主动命令执行、大规模重构、推测性代码适配或自动修复错误。
developmentcode-reviewresearch
ai-research-reproduction
lllllllama
RigorPilot 复现模式编排器,用于README优先的深度学习仓库复现。当用户需要端到端、最小可信流程时使用,该流程先读取仓库,选择最小的文档化推理或评估目标,协调输入、设置、可信执行、可选的可信训练、可选的仓库分析以及可选的论文差距解决,强制执行保守补丁规则,记录证据、假设偏差和人工决策点……
researchdevelopmentdocument
explore-code
lllllllama
We need to translate the given text from English to Simplified Chinese. The instruction says: preserve product names, protocol names, URLs, numbers, and technical terms. Do not add claims, explanations, Markdown, bullets, links, labels, prefixes, or extra commentary. Also, do not include the name "explore-code" unless it appears in the source text. The source text does not contain "explore-code" explicitly, so we don't include it. The text is a description of a skill. We need to translate it accurately, keeping technical terms like "LoRA", "adapter layers", "backbone", "head", "rollback-aware records", "explore_outputs/", "current_research". Also note "Rigor Improve implementation leaf skill" - likely a proper name or technical term? Possibly "Rigor Improve" is part of a skill name? But instruction says preserve product names, so we keep "Rigor Improve" as is? Or translate? The instruction says "preserve product names, protocol names, URLs, numbers, and technical terms
developmentresearchcode-review
minimal-run-and-audit
lllllllama
We need to translate the given English text into Simplified Chinese. The instruction says: "Translate only the text inside <text>. Do not include the name unless it appears in the source text." The name "minimal-run-and-audit" is not inside the <text> tags, so we should not include it. The text is a description of a skill. We need to preserve product names, protocol names, URLs, numbers, technical terms. The text includes "Rigor Run", "README-first", "repro_outputs/", "patch notes". These should be preserved as is or translated appropriately? "Rigor Run" might be a product name? It says "Rigor Run skill" - likely a proper name. "README-first" is a phrase. "repro_outputs/" is a directory path. "patch notes" is a common term. We should keep technical terms like "smoke test", "inference", "evaluation", "repository", "training execution", "repo intake", "paper lookup", "target selection", "end-to-end orche
developmenttestingcode-review
env-and-assets-bootstrap
lllllllama
用于README优先的深度学习仓库复现的Rigor Setup技能。当任务明确需要为README文档化的仓库在运行前准备保守的conda优先环境、检查点和数据集路径假设、缓存位置提示以及设置说明时使用。不用于仓库扫描、完整编排、论文解读、最终运行报告或与特定复现目标无关的通用环境设置。
developmentdevops
explore-run
lllllllama
We need to translate the given text from English to Simplified Chinese. The text describes a skill called "explore-run" (name to preserve). The instruction says to translate only the text inside <text>, and not include the name unless it appears in the source text. The name "explore-run" does appear in the source text? Actually, the source text starts with "Rigor Improve / Rigor Explore run leaf skill..." - here "run" is part of the phrase, but the name to preserve is "explore-run". The source text does not contain "explore-run" as a standalone term; it has "Rigor Explore run leaf skill". However, the instruction says "Name to preserve: explore-run". I think we should not add the name if it's not in the source. The source has "Rigor Explore run" which might be a compound. But the instruction says "Do not include the name unless it appears in the source text." So we only translate what's there. The name "explore-run" is not literally in the source as a
researchdevelopmentdata-analysis
safe-debug
lllllllama
用于深度学习研究工作的严格调试/严格审计技能。当用户粘贴回溯信息、终端错误、CUDA内存不足、检查点加载失败、形状不匹配、NaN损失症状或训练失败,并希望在打补丁前进行保守诊断,且调试修复与研究贡献明确分离时使用。不适用于大规模重构、推测性适配、自动探索性修补或常规仓库熟悉。
developmenttestingcode-review