wanshuiyin/Auto-claude-code-research-in-sleep

这个仓库里有 189 个技能，GitHub 星标 ★ 16,209。

ablation-plannerUse when main results pass result-to-claim (claim_supported=yes or partial) and
alphaxivQuick single-paper lookup via AlphaXiv LLM-optimized summaries with tiered sourc
analyze-resultsAnalyze ML experiment results, compute statistics, generate comparison tables an
arxivSearch, download, and summarize academic papers from arXiv. Use when user says "
auto-paper-improvement-loopAutonomously improve a generated paper via GPT-6-Astra xhigh review → implement
auto-review-loopAutonomous multi-round research review loop. In Copilot CLI it defaults to the n
auto-review-loop-llmAutonomous research review loop using any OpenAI-compatible LLM API. Configure v
auto-review-loop-minimaxAutonomous multi-round research review loop using MiniMax API. Use when you want
citation-auditZero-context verification that every bibliographic entry in the paper is real, c
claims-draftingDraft patent claims for an invention. Use when user says \"撰写权利要求\", \"draft cla
comm-lit-reviewCommunications-domain literature review with Claude-style knowledge-base-first r
deepxivSearch and progressively read open-access academic papers through DeepXiv. Use w
dse-loopAutonomous design space exploration loop for computer architecture and EDA. Runs
embodiment-descriptionWrite detailed embodiment descriptions for patent specifications. Use when user
exa-searchAI-powered web search via Exa with content extraction. Use when user says "exa s
experiment-auditAudit experiment integrity before claiming results. Uses cross-model review (ext
experiment-bridgeWorkflow 1.5: Bridge between idea discovery and auto review. Reads EXPERIMENT_PL
experiment-planTurn a refined research proposal or method idea into a detailed, claim-driven ex
experiment-queueSSH job queue for multi-seed/multi-config ML experiments with OOM-aware retry, s
feishu-notifySend notifications to Feishu/Lark. Internal utility used by other skills, or man
figure-descriptionProcess user-provided patent figures and generate formal drawing descriptions. U
figure-specGenerate deterministic publication-quality architecture, workflow, and pipeline
formula-derivationStructures and derives research formulas when the user wants to 推导公式, build a th
gemini-searchSearch research papers via Gemini for broad literature discovery. Use when user
grant-proposalDraft a structured grant proposal from research ideas and literature. Supports K
idea-creatorGenerate and rank research ideas given a broad direction. Use when user says "找i
idea-discoveryWorkflow 1: Full idea discovery pipeline to go from a broad research direction t
idea-discovery-robotWorkflow 1 adaptation for robotics and embodied AI. Orchestrates robotics-aware
integrity-forensicsRun the Anti-Autoresearch integrity-forensics sweep (span-anchored evidence ledg
interview-cheatsheetGenerate a long-form Chinese interview-prep cheat sheet on a specific ML/LLM top
invention-structuringStructure a raw invention idea into a formal invention disclosure. Use when user
jurisdiction-formatCompile patent application into jurisdiction-specific filing format. Use when us
kill-argumentTwo-thread adversarial review: a fresh reviewer constructs the strongest 200-wor
mermaid-diagramGenerate Mermaid diagrams from user requirements. Supports flowcharts, sequence
meta-applyPrivileged applier that LANDS meta-optimize / corpus-audit patches the user appr
meta-optimizeAnalyze ARIS usage logs and propose optimizations to SKILL.md files, reviewer pr
monitor-experimentMonitor running experiments, check progress, collect results. Use when user says
novelty-checkVerify research idea novelty against recent literature. Use when user says "查新",
openalexSearch academic papers via OpenAlex API for open citation data, institutional af
overleaf-syncTwo-way sync between a local paper directory and an Overleaf project, so ARIS au
paper-claim-auditZero-context verification that every number, comparison, and scope claim in the
paper-compileCompile LaTeX paper to PDF, fix errors, and verify output. Use when user says \"
paper-figureGenerate publication-quality figures and tables from experiment results. Use whe
paper-illustrationGenerate publication-quality AI illustrations for academic papers using Gemini i
paper-illustration-image2Generate publication-quality academic illustrations through a local Codex app-se
paper-planGenerate a structured paper outline from review conclusions and experiment resul
paper-posterDEPRECATED — superseded by /paper-poster-html. Kept only as a redirect for muscl
paper-poster-htmlDEFAULT poster pipeline — build an academic conference poster (ICML/NeurIPS/ICLR
paper-slidesGenerate conference presentation slides (beamer LaTeX → PDF + editable PPTX) fro
paper-talkEnd-to-end conference talk pipeline: paper → slide outline → Beamer + PPTX → per
paper-writeDraft LaTeX paper section by section from an outline. Use when user says \"写论文\"
paper-writingWorkflow 3: Full paper writing pipeline that goes from a narrative report to a p
patent-novelty-checkAssess patent novelty and non-obviousness against prior art. Use when user says
patent-pipelineFull patent drafting pipeline from invention description to jurisdiction-formatt
patent-reviewGet an external patent examiner review of a patent application. Use when user sa
pixel-artGenerate pixel art SVG illustrations for READMEs, docs, or slides. Use when user
prior-art-searchSearch patent databases and academic literature for prior art relevant to an inv
proof-checkerRigorous mathematical proof verification and fixing workflow. Reads a LaTeX proo
proof-orchestratorManage a stateful, run-directory-based proof project: continuation across runs,
proof-writerWrites rigorous mathematical proofs for ML/AI theory. Use when asked to prove a
qzcliManage GPU compute jobs on the Qizhi (启智) platform using qzcli — a kubectl-style
rebuttalWorkflow 4: Submission rebuttal pipeline. Parses external reviews, enforces cove
render-htmlRender an ARIS Markdown / JSON artifact (IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT
research-implement-featureBuild a working artifact from a plain \"implement X for me\" request: a running
research-litSearch and analyze research papers, find related work, summarize key ideas. Use
research-pipelineFull end-to-end research pipeline: from a broad research direction through idea
research-refineTurn a vague research direction into a problem-anchored, elegant, frontier-aware
research-refine-pipelineRun an end-to-end workflow that chains `research-refine` and `experiment-plan`.
research-reviewGet a deep critical review of research from an external reviewer backend (Codex
research-wikiPersistent research knowledge base that accumulates papers, ideas, experiments,
resubmit-pipelineWorkflow 5: orchestrate a text-only resubmit of a polished paper to a different
result-to-claimUse when experiments complete to judge what claims the results support, what the
run-experimentDeploy and run ML experiments on local, remote, Vast.ai, or Modal serverless GPU
semantic-scholarSearch published venue papers (IEEE, ACM, Springer, etc.) via Semantic Scholar A
serverless-modalRun GPU workloads on Modal — training, fine-tuning, inference, batch processing.
ablation-plannerUse when main results pass result-to-claim (`claim_supported = yes` or `partial`
alphaxivQuick single-paper lookup via AlphaXiv LLM-optimized summaries with tiered sourc
analyze-resultsAnalyze ML experiment results, compute statistics, generate comparison tables an
arxivSearch, download, and summarize academic papers from arXiv. Use when user says \
auto-paper-improvement-loopAutonomously improve a generated paper via GPT-6-Astra xhigh review → implement
auto-review-loopAutonomous multi-round research review loop. Repeatedly reviews using a secondar
auto-review-loop-llmAutonomous research review loop using any OpenAI-compatible LLM API. Configure v
auto-review-loop-minimaxAutonomous multi-round research review loop using MiniMax API. Use when you want
citation-auditZero-context verification that every bibliographic entry in the paper is real, c
claims-draftingDraft patent claims for an invention. Use when user says \"撰写权利要求\", \"draft cla
comm-lit-reviewCommunications-domain literature review with Claude-style knowledge-base-first r
deepxivSearch and progressively read open-access academic papers through DeepXiv. Use w
dse-loopAutonomous design space exploration loop for computer architecture and EDA. Runs
embodiment-descriptionWrite detailed embodiment descriptions for patent specifications. Use when user
exa-searchAI-powered web search via Exa with content extraction. Use when user says "exa s
experiment-auditAudit experiment integrity before claiming results. Uses fresh-agent GPT-6-Astra
experiment-bridgeWorkflow 1.5: Bridge between idea discovery and auto review. Reads EXPERIMENT_PL
experiment-planTurn a refined research proposal or method idea into a detailed, claim-driven ex
experiment-queueSSH job queue for multi-seed/multi-config ML experiments with OOM-aware retry, s
feishu-notifySend notifications to Feishu/Lark. Internal utility used by other skills, or man
figure-descriptionProcess user-provided patent figures and generate formal drawing descriptions. U
figure-specGenerate deterministic publication-quality architecture, workflow, and pipeline
formula-derivationStructures and derives research formulas when the user wants to 推导公式, build a th
gemini-searchSearch research papers via Gemini for broad literature discovery. Use when user
grant-proposalDraft a structured grant proposal from research ideas and literature. Supports K
idea-creatorGenerate and rank research ideas given a broad direction. Use when user says \"\
idea-discoveryWorkflow 1: Full idea discovery pipeline to go from a broad research direction t
idea-discovery-robotWorkflow 1 adaptation for robotics and embodied AI. Orchestrates robotics-aware
integrity-forensicsRun the Anti-Autoresearch integrity-forensics DETERMINISTIC slice (numeric core
interview-cheatsheetGenerate a long-form Chinese interview-prep cheat sheet on a specific ML/LLM top
invention-structuringStructure a raw invention idea into a formal invention disclosure. Use when user
jurisdiction-formatCompile patent application into jurisdiction-specific filing format. Use when us
kill-argumentTwo-thread adversarial review: a fresh reviewer constructs the strongest 200-wor
mermaid-diagramGenerate Mermaid diagrams from user requirements. Save .mmd and .md files to fig
meta-applyPrivileged applier that LANDS meta-optimize / corpus-audit patches the user appr
meta-optimizeAnalyze ARIS usage logs and propose optimizations to SKILL.md files, reviewer pr
monitor-experimentMonitor running experiments, check progress, collect results. Use when user says
novelty-checkVerify research idea novelty against recent literature. Use when user says \"\u6
openalexSearch academic papers via OpenAlex API for open citation data, institutional af
overleaf-syncTwo-way sync between a local paper directory and an Overleaf project, so ARIS au
paper-claim-auditZero-context verification that every number, comparison, and scope claim in the
paper-compileCompile LaTeX paper to PDF, fix errors, and verify output. Use when user says \\
paper-figureGenerate publication-quality figures and tables from experiment results. Use whe
paper-illustrationGenerate publication-quality AI illustrations for academic papers using Gemini i
paper-illustration-image2Generate publication-quality academic illustrations through a local Codex app-se
paper-planGenerate a structured paper outline from review conclusions and experiment resul
paper-posterDEPRECATED — superseded by /paper-poster-html. Kept only as a redirect for muscl
paper-poster-htmlDEFAULT poster pipeline — build an academic conference poster (ICML/NeurIPS/ICLR
paper-slidesGenerate conference presentation slides (beamer LaTeX → PDF + editable PPTX) fro
paper-talkEnd-to-end conference talk pipeline: paper → slide outline → Beamer + PPTX → per
paper-writeDraft LaTeX paper section by section from an outline. Use when user says \\\"\u5
paper-writingWorkflow 3: Full paper writing pipeline that goes from a narrative report to a p
patent-novelty-checkAssess patent novelty and non-obviousness against prior art. Use when user says
patent-pipelineFull patent drafting pipeline from invention description to jurisdiction-formatt
patent-reviewGet an external patent examiner review of a patent application. Use when user sa
pixel-artGenerate pixel art SVG illustrations for READMEs, docs, or slides. Use when user
prior-art-searchSearch patent databases and academic literature for prior art relevant to an inv
proof-checkerRigorous mathematical proof verification and fixing workflow. Reads a LaTeX proo
proof-orchestratorManage a stateful, run-directory-based proof project with Codex: continuation ac
proof-writerWrites rigorous mathematical proofs for ML/AI theory. Use when asked to prove a
qzcliManage GPU compute jobs on the Qizhi (启智) platform using qzcli — a kubectl-style
rebuttalWorkflow 4: Submission rebuttal pipeline. Parses external reviews, enforces cove
render-htmlRender an ARIS Markdown / JSON artifact (IDEA_REPORT, AUTO_REVIEW, KILL_ARGUMENT
research-implement-featureBuild a working artifact from a plain \"implement X for me\" request: a running
research-litSearch and analyze research papers, find related work, summarize key ideas. Use
research-pipelineFull end-to-end research pipeline: from a broad research direction through idea
research-refineTurn a vague research direction into a problem-anchored, elegant, frontier-aware
research-refine-pipelineRun an end-to-end workflow that chains `research-refine` and `experiment-plan`.
research-reviewGet a deep critical review of research from GPT using a secondary Codex agent. U
research-wikiPersistent research knowledge base that accumulates papers, ideas, experiments,
resubmit-pipelineWorkflow 5: orchestrate a text-only resubmit of a polished paper to a different
result-to-claimUse when experiments complete to judge what claims the results support, what the
run-experimentDeploy and run ML experiments on local or remote GPU servers. Use when user says
semantic-scholarSearch published venue papers (IEEE, ACM, Springer, etc.) via Semantic Scholar A
serverless-modalRun GPU workloads on Modal — training, fine-tuning, inference, batch processing.
slides-polishPer-page Codex review + targeted python-pptx / Beamer fixes for academic talk sl
specification-writingWrite the full patent specification from claims and invention disclosure. Use wh
system-profileProfile a target (script, process, GPU, memory, interconnect) for performance an
training-checkInteractively monitor training metrics from the current Codex session, periodica
vast-gpuRent, manage, and destroy GPU instances on vast.ai. Use when user says \"rent gp
web-debug-searchSearch GitHub, Stack Exchange, Chinese technical communities, official documenta
wiki-enrichFill in the per-paper TODO sections of research-wiki/papers/<slug>.md pages that
writing-systems-papersParagraph-level structural blueprint for 10-12 page systems papers targeting OSD
auto-paper-improvement-loopAutonomously improve a generated paper via Claude review through claude-review M
auto-review-loopAutonomous multi-round research review loop. Repeatedly reviews using Claude Cod
novelty-checkVerify research idea novelty against recent literature. Use when user says \"查新\
paper-figureGenerate publication-quality figures and tables from experiment results. Use whe
paper-planGenerate a structured paper outline from review conclusions and experiment resul
paper-writeDraft LaTeX paper section by section from an outline. Use when user says \"写论文\"
research-refineTurn a vague research direction into a problem-anchored, elegant, frontier-aware
research-reviewGet a deep critical review of research from Claude via claude-review MCP. Use wh
auto-paper-improvement-loopAutonomously improve a generated paper via Gemini review through gemini-review M
auto-review-loopAutonomous multi-round research review loop. Repeatedly reviews using Gemini via
grant-proposalDraft a structured grant proposal from research ideas and literature. Supports K
idea-creatorGenerate and rank research ideas given a broad direction. Use when user says \"\
idea-discoveryWorkflow 1: Full idea discovery pipeline. Orchestrates research-lit \u2192 idea-
idea-discovery-robotWorkflow 1 adaptation for robotics and embodied AI. Orchestrates robotics-aware
novelty-checkVerify research idea novelty against recent literature. Use when user says \"查新\
paper-figureGenerate publication-quality figures and tables from experiment results. Use whe
paper-planGenerate a structured paper outline from review conclusions and experiment resul
paper-poster-htmlDEFAULT poster pipeline — build an academic conference poster (ICML/NeurIPS/ICLR
paper-slidesGenerate conference presentation slides (beamer LaTeX → PDF + editable PPTX) fro
paper-writeDraft LaTeX paper section by section from an outline. Use when user says \"写论文\"
paper-writingWorkflow 3: Full paper writing pipeline. Orchestrates paper-plan \u2192 paper-fi
research-refineTurn a vague research direction into a problem-anchored, elegant, frontier-aware
research-reviewGet a deep critical review of research from Gemini via gemini-review MCP. Use wh
slides-polishPer-page Codex review + targeted python-pptx / Beamer fixes for academic talk sl
specification-writingWrite the full patent specification from claims and invention disclosure. Use wh
system-profileProfile a target (script, process, GPU, memory, interconnect) for performance an
training-checkPeriodically check WandB metrics during training to catch problems early (NaN, l
vast-gpuRent, manage, and destroy GPU instances on vast.ai. Use when user says \"rent gp
web-debug-searchSearch GitHub, Stack Exchange, Chinese technical communities, official documenta
wiki-enrichFill in the per-paper TODO sections of research-wiki/papers/<slug>.md pages that
writing-systems-papersParagraph-level structural blueprint for 10-12 page systems papers targeting OSD

想一次拿到这个仓库的全部技能？

本站把开放许可（MIT / Apache 等）的仓库按整仓打包整理到网盘，点一下转存到你自己的网盘。许可未声明的仓库只给原始仓库链接，不打包。

去原始仓库