AI research workflow framework for project setup, GPU experiments, agent collaboration, Feishu remote control, and reproducible paper pipelines.
This repository also configures its own agents. See what Labline tells them →
Skill Claude Code
Use when main results pass result-to-claim (claimsupported=yes or partial) and ablation studies are needed for paper submission. Codex designs ablations from a reviewer's perspective, CC reviews feasibility and implements.
Skill Claude Code
Quick single-paper lookup via AlphaXiv LLM-optimized summaries with tiered source fallback. Use when user says "explain this paper", "summarize paper", pastes an arXiv/AlphaXiv URL, or provides a bare arXiv ID for quick understanding - not for broad literature search.
Skill Claude Code
Analyze ML experiment results, compute statistics, generate comparison tables and insights. Use when user says "analyze results", "compare", or needs to interpret experimental data.
Skill Claude Code
Search, download, and summarize academic papers from arXiv. Use when user says "search arxiv", "download paper", "fetch arxiv", "arxiv search", "get paper pdf", or wants to find and save papers from arXiv to the local paper library.
Skill Claude Code
A skill that reviews a generated research paper, applies the suggested fixes and recompiles it for two rounds. Recompiling means producing the updated document again after changes.
Skill Claude Code
Autonomous research review loop using any OpenAI-compatible LLM API. Configure via llm-chat MCP server or environment variables. Trigger with "auto review loop llm" or "llm review".
Skill Claude Code
Autonomous multi-round research review loop using MiniMax API. Use when you want to use MiniMax instead of Codex MCP for external review. Trigger with "auto review loop minimax" or "minimax review".
Skill Claude Code
Autonomous multi-round research review loop. Repeatedly reviews via Codex MCP, implements fixes, and re-reviews until positive assessment or max rounds reached. Use when user says "auto review loop", "review until it passes", or wants autonomous iterative improvement.
Skill Claude CodeCodex
Ultra-compressed communication mode. Cuts token usage 75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.
Skill Claude Code
Zero-context verification that every bibliographic entry in the paper is real, correctly attributed, and used in a context the cited paper actually supports. Uses a fresh cross-model reviewer with web/DBLP/arXiv lookup to catch hallucinated authors, wrong years, fabricated venues, version mismatches, and wrong-context…
Skill Claude Code
A patent-claim drafting workflow for writing the legal statements that define what an invention is intended to protect. It is used when creating claims for a patent application.
Skill Claude Code
A coding role that writes code, tests, and refactors existing code. It does not handle deployment, SSH access, or academic paper writing.
Skill Claude Code
Communications-domain literature review with Claude-style knowledge-base-first retrieval. Use when the task is about communications, wireless, networking, satellite/NTN, Wi-Fi, cellular, transport protocols, congestion control, routing, scheduling, MAC/PHY, rate adaptation, channel estimation, beamforming, or…
Skill Claude Code
Search and progressively read open-access academic papers through DeepXiv. Use when the user wants layered paper access, section-level reading, trending papers, or DeepXiv-backed literature retrieval.
Skill Claude Code
A deployment role that transfers files over SSH, starts training runs, monitors experiments, and collects their results without writing code. SSH is a secure way to connect to another computer over a network.
Skill Claude CodeCodex
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
Skill Claude Code
Autonomous design space exploration loop for computer architecture and EDA. Runs a program, analyzes results, tunes parameters, and iterates until objective is met or timeout. Use when user says "DSE", "design space exploration", "sweep parameters", "optimize", "find best config", or wants iterative parameter tuning.
Skill Claude Code
A writing tool for detailed embodiment sections in patent specifications. An embodiment explains one practical way an invention can be built or used.
Skill Claude Code
AI-powered web search via Exa with content extraction. Use when user says "exa search", "web search with content", "find similar pages", or needs broad web results beyond academic databases (arXiv, Semantic Scholar).
Skill Claude Code
An experiment-integrity checker that reviews completed experiments before you report their results. It looks for false reference answers, manipulated scores, made-up results, and experiments that cover too little.
Skill Claude Code
A workflow that turns an experiment plan in EXPERIMENTPLAN.md into running code, deploys it to a graphics processor (GPU), and gathers initial results.
Skill Claude Code
Turn a refined research proposal or method idea into a detailed, claim-driven experiment roadmap. Use after research-refine, or when the user asks for a detailed experiment plan, ablation matrix, evaluation protocol, run order, compute budget, or paper-ready validation that supports the core problem, novelty…
Skill Claude Code
An SSH-based queue for running many machine-learning experiments across different random seeds and configurations. It includes handling for out-of-memory failures, stale terminal sessions, and transitions between groups of jobs.
Skill Claude Code
A notification skill for sending messages to Feishu, also called Lark, a workplace communication app.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: