Look up any arxiv paper on alphaxiv.org to get a structured AI-generated overview. This is faster and more reliable than trying to read a raw PDF.
中文简介正在准备中。当前显示原文说明。
wandb/senpai372026年10月21日 更新
免费在 GitHub 公开
maintain-research-state
Maintain and publish the advisor-owned current research state, durable research-ideas syntheses, and dataset analysis. Use when evidence or human guidance changes the research focus or direction, durable research ideas should be recorded, or dataset understanding changes.
中文简介正在准备中。当前显示原文说明。
查看安装方法包含的文件（1）
- SKILL.md1.7 KB
SKILL.md（原文）
安装前可以查看交给 AI 代理的指令内容。
Maintain research state
Use the immutable program.md snapshot embedded in your system prompt. It defines
target-specific objectives, metrics, constraints, protected files, and any
baseline or research-log format.
Maintain these advisor-owned artifacts:
-
Keep
research/CURRENT_RESEARCH_STATE.mdas a living, pruned view of the current high-level focus, themes, hypotheses, experiments, and potential next directions. Do not use it as an archive or experiment-by-experiment log. Use this schema:# SENPAI Research State - <current date and time> - <most recent research direction from human researcher team> - <current research focus and themes> - <list of potential next research directions and themes> -
Optionally record durable research or round-planning synthesis in
research/RESEARCH_IDEAS_<YYYY-MM-DD_HH:MM>.md. -
Save rigorous dataset analysis and durable future dataset insights in
research/DATASET_ANALYSIS.md.
Commit only advisor-owned changes. Read the current remote advisor-branch SHA
and the local commit SHA, then call publish_advisor_branch:
{
"remote_branch_sha_before_push": "REMOTE_SHA_BEFORE_PUSH",
"local_commit_sha": "LOCAL_COMMIT_SHA"
}
Do not publish the advisor branch with git push.
评价
还没有评价。欢迎分享你的使用感受。
评论
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
同一仓库的技能
Operator-side analysis of historical ML experiment PRs in Senpai research tracks. Use this skill whenever the user asks to: analyze experiments, categorize PRs, bucket experiments, summarize what's been tried, understand experiment history, review merged vs closed results, or asks "what experiments have we run / worked / failed". Also triggers for: "pull the latest experiments", "what's been tried so far", "category breakdown of PRs", "which experiments succeeded", "noam track analysis". When a branch name is mentioned (e.g. "noam branch", "on the noam branch"), pass it as the base branch to scope the fetch to just those PRs.
中文简介正在准备中。当前显示原文说明。
wandb/senpai372026年10月21日 更新
Create a typed assignment branch and draft PR for one student. Use when the advisor has a concrete hypothesis and the student has no open `status:wip` or `status:review` assignment.
中文简介正在准备中。当前显示原文说明。
wandb/senpai372026年10月21日 更新
Create or improve a Senpai target repository's program.md. Use this skill whenever the user wants to point Senpai at a fresh ML or research target repository, define the research objective, primary metric, benchmark contract, allowed edit boundaries, W&B reporting contract, or prepare a repo for autonomous advisor/student experiment loops.
中文简介正在准备中。当前显示原文说明。
wandb/senpai372026年10月21日 更新
Open GitHub Issues for human input and respond to the researcher team. Use this skill whenever you need to handle a human_issue event, respond to human issues, ask humans a question, or check team communications. Also triggers for: "any human messages?", "check issues", "respond to humans".
中文简介正在准备中。当前显示原文说明。
wandb/senpai372026年10月21日 更新
Choose, configure, launch, and collect bounded subagents for research and engineering decisions. Root agents and delegation-capable subagents should read this before delegating: every task requires an explicit model tier, and high-leverage work such as research ideation, round planning, plateau pivots, large research reviews, hard optimization, disputed evidence, and expensive experiment portfolios requires frontier judgment.
中文简介正在准备中。当前显示原文说明。
wandb/senpai372026年10月21日 更新