Paper tutorial briefing
Skill StaryMoon/ai-researcher-skills/skills/paper-tutorial-briefing
Practical skill pack for AI research agents: paper briefing, experiment planning, rebuttal drafting, radar maintenance, and GitHub upkeep.
npx -y skills add StaryMoon/ai-researcher-skills --skill paper-tutorial-briefingAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.2 KB, as published. Nobody here has run it
Paper Tutorial Briefing
Purpose
Turn an AI paper into a rigorous but readable tutorial briefing. The goal is not to paraphrase the abstract; the goal is to help a researcher understand what the paper changes, what evidence supports it, and what should be tried locally next.
When To Use
Use this skill when the user gives a paper PDF, arXiv link, title, project page, README, or local paper note and asks for a technical explanation, tutorial, daily briefing entry, or research direction.
Inputs
- Paper PDF, URL, title, or extracted text.
- Optional code repository or project page.
- Optional local research context, baseline, metric, or dataset.
- Optional target audience such as beginner, reviewer, implementer, or project maintainer.
Workflow
- Identify the paper metadata: title, venue or preprint status, authors if available, task, model family, and main artifact.
- Produce a one-sentence takeaway that states the technical move, not the marketing claim.
- Extract the contribution stack: problem setting, key mechanism, training or inference recipe, evidence, and stated limitations.
- Build an evidence map. Separate paper-reported metrics, qualitative figures, ablations, and any local evidence.
- Explain the method with one concrete implementation mental model: tensors, modules, prompts, losses, memory, data flow, or evaluation loop.
- Identify reproduction risk: missing details, compute assumptions, hidden preprocessing, dataset ambiguity, or metric mismatch.
- End with next actions: shortest faithful experiment, stop rule, and what result would justify adding it to a repo or briefing.
Output
Use this structure unless the user asks otherwise:
One-Sentence TakeawayWhy This Paper ExistsCore MechanismEvidence MapImplementation NotesReproduction RisksNext Local Actions
Quality Gates
- Do not claim results were reproduced unless local commands or logs prove it.
- Mark uncertainty explicitly:
paper claim,local evidence, orinference. - Prefer named metrics, datasets, and ablations over vague praise.
- If the paper text is unavailable, say what is missing and base the briefing only on accessible sources.