Summarization Agent Skills

Maps where a movement leader's content and relationships live, producing a fragmentation report from existing research files.

Takes pre-computed differential expression results from DESeq2, edgeR, limma, or PyDESeq2 and produces a structured, publication-ready summary with ranked gene lists, biological themes, and key observations.

Runs Stanford's STORM pipeline to produce Wikipedia-quality research articles with citations, using Claude models and a search engine API.

Explains cellular structures, membrane transport, energetics, signaling, and division, connecting molecular events to organismal function.

Merges multiple research, analysis, or brainstorming inputs into a single coherent synthesis with key insights and a recommended action.

Embody the theologian Hans Kung to analyze institutions, bridge divides, and apply the Global Ethic framework.

Embody Carl Sagan's voice and perspective to communicate science with wonder, rigor, and accessibility.

规划、撰写、重构、润色、压缩、扩写、收口和审校中文医学文本，并为持续项目建立写作主控索引、内容型总纲、全文一致性账本、章节任务书、证据账本、决策记录和最终交付物清单。支持 Grant（医学基金标书、项目申请、研究目标、任务分解、技术路线、进度与甘特图、项目管理、风险分析、创新性、预期效益、研究基础和可行性论证）与 Journal（毕业论文、期刊论文、综述、病例报告、指南解读和学术报告），自动识别或联合调用全文建构、局部构篇、语言精修、结构审计、语言审计及项目收口。

Reviews an academic paper using the NeurIPS review form with three reviewer personas, ensemble scoring, and reflection refinement. Extracts text from PDF, runs structured review, and outputs actionable feedback.

Benchmarks large audio-language models against adversarial audio attacks using the CAA dataset, computing WER, ROUGE-L, cosine similarity, and coherence scores to assess robustness in conversational settings.

Benchmarks LLM bias by scoring responses to automatically generated open-ended questions across sensitive attributes, producing a composite fitness score from 0 to 5.

Draft publication-ready ML/AI papers for top conferences (NeurIPS, ICML, ICLR, ACL, AAAI, COLM) by exploring research repositories, structuring arguments, verifying citations via APIs, and formatting submissions.

Applies cognitive science frameworks for creative thinking to CS and AI research ideation, using combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded strategies.

Create refined user personas from research data with demographics, goals, frustrations, and behavioral patterns for product and UX design decisions.

Transform user interview transcripts into structured, actionable insights with key themes, quotes, and action items.

Fans out web searches, fetches sources, adversarially verifies claims, and synthesizes a cited research report.

Analyzes why a piece of content resonates with audiences using five classic communication theories, identifies the emotional core and effective stance, and outputs discussion directions for a chatroom.

Finds and exports podcast episode transcripts as cleaned TXT files from YouTube URLs, episode webpages, Apple Podcasts, X/Twitter links, direct audio URLs, or plain titles, with a deterministic decision tree and optional local ASR fallback.

Rekonstruiert den zeitlichen Ablauf eines Datenschutzvorfalls für die DSGVO-Meldung, prüft Fristen, Form, Zuständigkeit und Rechtsweg und liefert eine Fristen- und Risikoampel mit Sofortmaßnahmen.

Explains Solana's architecture and protocol internals, covering the SVM execution engine, account model, consensus, transactions, validator economics, data layer, development tooling, and token extensions using Helius blog posts, SIMDs, and Agave/Firedancer source code.

Performs AI-powered web searches with real-time information using Perplexity models via LiteLLM and OpenRouter, providing grounded answers with source citations.

Orchestrates structured experiments end-to-end: initialize, plan, run, and evolve with reproducibility tracking and hypothesis validation.

Verifies that every number, comparison, and scope claim in a research paper matches raw result files, using a fresh cross-model reviewer with no prior context to prevent confirmation bias.

Explains biomedical concepts, vital signs, biological signals, medical device classifications, and healthcare data standards for technical teams.

Identifies and prioritizes knowledge gaps, then systematically consults workspace files, past conversations, documentation, and web search to fill them with citable, confidence-rated answers.

Monitors trending topics across 35+ platforms, aggregates trends, analyzes sentiment, and provides real-time notifications via MCP tools.

Fetches and summarizes the latest ORF news in German, focusing on Austrian and international politics, and generates a themed studio image.

Formalizes narrative technical documents into structured mathematical problem specifications with variables, constraints, objectives, and uncertainty, citing evidence and flagging missing information.

Converts narrative technical documents into structured mathematical problem specifications with variables, constraints, objectives, and uncertainty.

Records research provenance as a post-task epilogue, scanning conversation history to extract decisions, experiments, dead ends, claims, heuristics, and pivots, and writing them into the ara/ directory with provenance tags.

Adopts the perspective of Naval Ravikant to answer questions using his mental models, decision heuristics, and core concepts like leverage, specific knowledge, and desire management.

Verifies claims, articles, screenshots, and URLs through source-grounded analysis with an evidence ledger, source credibility evaluation, and manipulation detection. Supports quick checks, full fact-check cards, two-source comparisons, and prebunking in multiple languages and policy contexts.

Erstellt Entwürfe für Industriekunden im Energierecht: prüft Sachverhalt, Normen und Beweise und liefert Anträge mit Begründung und Anlagenlogik.

Provides a general standard operating procedure for handling requests related to news reports about Chinese leadership visits, including title generation and step-by-step execution guidance.

Verifies a claim by finding primary sources and rates the evidence as supported, mixed, unsupported, or unverifiable, with citations and what would change the verdict.

智能路由中心，自动识别用户意图并路由到相应技能，提供专业化的深度研究体验，支持多Agent并行调研与监控。