Tracks every paper, study, dataset, and source that has been shared, recommended, or discussed across research email threads — building a structured reading list with context on…
Interpret figures in academic papers and their captions when the input is a PDF-to-Markdown document with page markers and image links, producing a structured Markdown report for…
Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral).
Academic literature orientation skill that searches papers via Consensus, builds a strategic search plan using PICO (default) or SPIDER / Decomposition / hybrid as fallbacks, and…
Trace the intellectual lineage of the 3 most-mentioned concepts in a research corpus — who first proposed each idea, who challenged it, who refined it, and where consensus stands…
Surface every conflict in a research corpus — every point where 2+ author groups reach opposite conclusions, with positions, papers, and the methodological reasons for the…
Identify 5 research questions the corpus does NOT fully answer — and for each: why the gap persists, which paper comes closest, what methodology would fill it.
Literature review intake — when the user uploads multiple research papers, produce a corpus-level inventory before any deep questions: paper-by-paper one-liner (author, year, key…
Build a structured knowledge map of a research field — central thesis, 3–5 supporting findings, 2–3 active debates, 1–2 frontier questions, 3 starter papers with reasons.
Compress everything known about a research field into a 5-minute briefing for an educated non-expert — three sentences: what's proven, what's unknown, what's the real-world…
Conservation checks from Little's Law (`L = λW`) and queueing decisions: measurement boundaries, service demand versus residence time, utilisation curves, thread pool and executor…
Produce a deep, multi-source, cited research brief on a topic from live web data using Bright Data's Discover API (intent-ranked web search + parsed page content).
Update or version a PKB document that evolves over time — supersede the old version with a new one while preserving history.
Use when targeting Living Reviews in Relativity (LRR) or deciding whether an invited, continuously updated authoritative review in relativity and gravitation fits this…
Paper reader for non-academics. Takes a paper and extracts its ideas for personal use. Focuses on understanding, not academic critique.
Paper workflow: read papers + cast cards in one go. Takes one or more arxiv links, paper URLs, PDFs, or paper names.
论文倒读法：给一篇论文，递归找出它批判和改进的前序论文（最多5层），再找它之后的最新进展，从源头正向讲述问题演化史。以问题为轴，费曼式讲解每篇论文看到的问题和解法创新。Use when user shares a paper and wants to understand its intellectual lineage, citation chain,…
Deep travel research workflow for museums and ancient architecture. Input a city name, auto-generates structured knowledge document (org-mode) + portable reference cards (PNG).
Deconstructs academic papers like an X-ray machine, extracting core contributions, critical assumptions, and napkin-worthy insights from research papers.
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable — from…
Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support — from Orchestra Research
Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models.
Build and maintain a Karpathy-style LLM knowledge base — a persistent, compounding wiki of interlinked markdown files that an LLM compiles from raw sources and maintains over…
Implement, configure, and maintain the LLM Lab for testing skills across multiple LLMs, comparing their results, and building autonomous pipelines.
Co-create a personal investment-research LLM Wiki (Andrej Karpathy's pattern) where the user's OWN analysis framework becomes a living CLAUDE.md — by interviewing them, NOT by…
Generate and update HTML dashboards for LLM usage (Claude, Gemini, Kiro, VS code, Cline, etc). Use when the user wants to visualize their AI coding assistant usage statistics,…
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic r — from…
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic r — from…
Research standards/patterns via MCP Ref. Generates Standards Research for Story Technical Notes subsection. Reusable worker. — from science/science-misc
Research standards/patterns via MCP Ref. Generates Standards Research for Story Technical Notes subsection. Reusable worker. — from majiayu000/claude-skill-registry
Research standards/patterns via MCP Ref. Generates Standards Research for Story Technical Notes subsection. Reusable worker. — from majiayu000/claude-skill-registry
Researches standards and patterns via MCP Ref for Story Technical Notes. Use when Story needs standards research before task generation.
Research best practices via MCP Ref/Context7/WebSearch and create documentation (guide/manual/ADR/research). Single research, multiple output types. — from science/science-misc
Research best practices via MCP Ref/Context7/WebSearch and create documentation (guide/manual/ADR/research). Single research, multiple output types.
Researches best practices and creates documentation (guide, manual, ADR, or research). Use when project needs standards-based documentation for a topic.
Indexes and queries project research graphs backed by hex-research MCP. Use for hypotheses, goals, benchmark runs, evidence depth, derived goal metrics, lineage, generated…
Use when validating Stories, plans, or context through the evaluation platform with mandatory research, parallel evidence lanes, sequential merge, and bounded refinement.
Use when an evaluation run needs mandatory official-doc, MCP Ref, Context7, and current best-practice research with a structured research summary. — from science/research-methods
Use when an evaluation run needs mandatory official-doc, MCP Ref, Context7, and current best-practice research with a structured research summary.
Use when an evaluation coordinator needs normalized findings from target artifacts and research evidence.
Use when an evaluation run must merge research, findings, documentation, and repair outputs into one verified result.
Use when coordinating story quality evaluation with mandatory research, worker summaries, agent review, regression evidence, and bounded refinement.
Researches real-world problems, competitor solutions, and customer complaints before test planning. Posts findings as Linear comment for ln-512 and ln-513.
Researches real-world problems, competitor solutions, and customer complaints before test planning. Posts findings as Linear comment for ln-512 and ln-513.
Researches real-world problems, competitor solutions, and customer complaints for a feature domain. Use before test planning to ground tests in actual user pain points.
Use when auditing project documentation through the evaluation platform with mandatory research, coordinated audit workers, and structured summaries.
Use when auditing the codebase through the evaluation platform with mandatory research, coordinated domain audit workers, and structured summaries.
Use when auditing architectural patterns through the evaluation platform with mandatory best-practice research, coordinated pattern workers, and structured summaries.
Use when auditing architectural patterns through the evaluation platform with mandatory best-practice research, coordinated pattern workers, and structured summaries.
Use when auditing persistence and runtime performance through the evaluation platform with mandatory research, coordinated data-layer workers, and structured summaries.
Coordinates 3 specialized audit workers (query efficiency, transaction correctness, runtime performance).
Researches competitive benchmarks and generates optimization hypotheses for identified bottlenecks. Use after profiling.
Use when validating optimization plans through the evaluation platform with mandatory research, parallel agent evidence, sequential merge, and bounded refinement.
Use this bioinformatics data analysis skill to construct a database-driven lncRNA-mRNA regulatory network from target lncRNA and/or gene lists by projecting shared miRNA evidence…
Use when agents must debate, conference, deliberate, or reach consensus on a goal — competing positions argue and converge on one deliverable, adversarial review with synthesis,…
Surface and classify all open questions, unresolved contradictions, stale claims, and knowledge gaps from existing memory.
Create or extend an Obsidian-friendly markdown wiki scaffold in the current workspace. Use this when the user wants to build a wiki, set up a knowledge base, create a research…
Research workspace context, APIs, implementation options, or external evidence before planning. The terminal artifact is always a spec at specs/.md; a…
Develop, extend, and contribute to Lobster AI — the multi-agent bioinformatics engine. Use when working on Lobster codebase, creating agents/services, understanding architecture,…
Run bioinformatics analyses using Lobster AI - single-cell RNA-seq, bulk RNA-seq, literature mining, dataset discovery, quality control, and visualization.