Research & Data

Assesses the reliability of web sources and the claims they support. Use when you have a set of research claims plus a table of sources (S-ids with titles, page age and evidence labels) and must rate each claim High/Medium/Low, note corroboration, contradictions and unsupported claims, without having opened any page.

The rmfit process for global R-matrix fitting with AZURE2 -- start a campaign on any reaction, drive or resume one, use idle time while jobs run for input-file / data audits and literature review (sharded pyazr evaluation, bound-tightening polish, interference-sign rounds, level add/remove rounds, dataset-off diagnostics, ledger + STATUS.md), judge results, export for GUI review, and refine the method. Use whenever the task mentions rmfit, a campaign directory, STATUS.md, sign rounds, a tightening polish, a level-add test, fitting a new reaction "the rmfit way", or continuing a fit that a previous session left running on the cluster.

Turns assessed, source-rated research findings into a short, structured, citation-safe research brief for the end user. Use when you hold a research question, an Analyst summary, findings rated High/Medium/Low with source ids, contradictions, unsupported claims and a source table (S-ids), and must write the final Markdown answer with no tools, citing only [S#] ids and keeping the stated confidence of every finding intact.

Use when a question involves U.S. Department of Energy emergency orders under Federal Power Act section 202(c), for example whether a power plant (J.H. Campbell, Eddystone, H.A. Wagner, Centralia, R.M. Schahfer, F.B. Culley, Craig, Stanton) is under a DOE order to keep running, when its current order expires, how many times it has been renewed, which grid-emergency orders (PJM, ERCOT, Duke, SPP, ISO-NE, NYISO, Florida utilities) or island orders (Puerto Rico's PREPA, Guam) were in force on a date, or how many megawatts the plant orders cover. Ships an offline, standard-library lookup over a table of all 87 orders, extensions and amendments of May 2025 to 6 November 2026, transcribed from DOE's pages, with the table's as-of date and the traps in reading these orders.

Apply sceptical senior-analyst scrutiny to claims, case studies, statistics, vendor material, reports and recommendations, in either a short gut-check or a long detailed dissection. Use when the user asks to sanity-check, challenge, red-team or "think critically about" content, or pastes/links a case study, ROI or performance claim, or "independent" study and wants it assessed. Do not use for plain summarisation or for content the user has not asked to be evaluated.

Searches official scholarly APIs (arXiv, Semantic Scholar, Europe PMC, OpenAlex), reads the legal PDF, and writes a cited note from those papers. Use when the user wants scientific papers, a literature review, state of the art, evidence on a topic, a known DOI, or says "busca papers sobre X", even if they never use the word paper.

Generic patterns for orchestrating data analysis pipelines in R using GNU Make with file-target workflows and standardized script contracts

Build a page, app, game, quiz or AI tutor that students will use, or a tool that touches student data, so that it holds up in a real classroom. Use when a teacher, researcher or developer asks you to build or change anything learners will use, however small, and before it is shared with a class.

The Earnings Analyst. Reads a company's reported numbers straight from SEC EDGAR (XBRL company facts from its 10-Qs and 10-Ks, plus the latest earnings 8-K and its press release) and sums up what changed: revenue, gross, operating and net margins, net income and diluted EPS, quarter by quarter, each against the same quarter a year earlier, with the trailing twelve months. Reads an earnings call transcript when you paste one. Use when the user says "how did Tesla's earnings look", "analyze the earnings", "what changed this quarter", "is revenue growing", "check the 10-Q", "compare margins", "sum up the call", "does it report this week", or before trading a stock around its earnings.

News Monitor. Tracks breaking news on every stock you watch and flags what moves it: Yahoo Finance headlines per ticker, each company's 8-K filings on SEC EDGAR (earnings, CEO changes, deals, restatements, bankruptcy) and how far each stock is from yesterday's close. Shows only what is new since the last check, newest first, in New York time, tags each item (earnings, analyst, deal, halt, regulatory, macro...) and never repeats one. Use when the user says "any news on TSLA", "why is this moving", "what's the news on my watchlist", "morning briefing", "pre-market news", "monitor the news", "did anything break", "check the 8-Ks", or after /trader-scanner finds setups.

问卷设计与在线收集系统生成器。把一份问卷需求（录音口述、文字大纲、散乱题目）一键变成「Word 问卷 + 手机扫码答题网站 + 后台数据自动汇总看板」三件套，并产出可一键运行的完整系统。当用户说“设计一份问卷”“根据录音设计问卷”“做个调查问卷”“生成问卷”“在线问卷”“扫码答题的网站”“问卷数据汇总”“问卷后台”“问卷统计”“把问卷做成网页/网站”“老师们扫码填问卷”时使用。也适用于把已有题目整理成规范问卷或在线收集系统。不适用于非问卷类文档生成、单纯的问卷统计数据分析（无新问卷设计）、考试组卷（用 question-generator）。

Run IES-backed Radiance direct and reflected sensor illuminance calculations with numerical-stability checks for supported rectangular rooms and downlights. Use for electric-light photometric analysis.

Converts P&IDs (piping and instrumentation diagrams) and PFDs in PDF form, either vector or scanned, into a process knowledge graph covering equipment connectivity (upstream/downstream), instrument locations and measured variables, and control loops. Output is JSON plus a DuckDB file plus an LLM-ready Markdown summary. Use when the user wants to digitize, extract, analyze or build a knowledge graph or context layer from P&IDs, or asks what is upstream/downstream of a piece of equipment, which sensors measure what, or how a plant's control loops are structured.

Draft a structured preregistration document (OSF, AsPredicted, or AEA RCT Registry style) from a research spec or free-form study description. Output is a Markdown file with hypotheses, design, sampling plan, analysis plan, exclusions, and inference criteria — annotated with MUST / SHOULD / MAY clarity flags. Use when user says "preregister", "draft a preregistration", "OSF preregistration", "AsPredicted", "AEA RCT registry", "PAP", "preanalysis plan", or before launching an experiment / data co

Use when citing web pages as evidence, either in a markdown file (specs, design notes, RFCs, READMEs) or in a chat answer where the user asks for sources, references, citations, quotes, or proof they can check, or when a reader needs to see what a linked source actually says without opening it.

Checks a public website against core German and EU legal duties (Impressum/DDG, privacy policy and consent/TDDDG/DSGVO, HTTPS, prices/PAngV, order, cancellation and withdrawal buttons/BGB, dispute resolution/VSBG, accessibility/BFSG, DSA, AI Act, green claims/UWG) and helps fix every gap with text drafts, mock-ups and official sources. Checks one or several websites (separated by ;), detects the system they are built with (e.g. WordPress, Shopify, Wix) and writes layperson fix steps for it. Asks

Research a person's, team's, or organization's GitHub work over a year or another period, then create a source-backed vertical timeline and individually tailored Wrapped with an offline HTML export. Use for GitHub yearbooks, annual engineering retrospectives, and contribution wrap-ups.

查证和解读指定论文，将中文元数据、解读笔记和主题标签整理到对应的 Zotero 条目。用于按论文标题、DOI、论文链接或 Zotero 条目执行文献整理。

Finds video ads that are probably working for restaurants and food brands, using the free Meta Ad Library, and builds a short "swipe file" of likely winners to study. Use when the user asks to find winning ads, research competitors' Facebook or Instagram ads, see what ads restaurants are running, or build ad inspiration for a food business.

>- Formalize a mathematics research paper in Lean 4 with Mathlib, end to end, in any area of mathematics: inventory the paper, state each result faithfully, prove the paper's own results and then the results it cites, make an existing Lean draft (often AI-written and never compiled) build with no sorry or axiom, audit the paper against the formalization in a report, explore which follow-up papers it could lead to, and package the project for the Palomar registry (Challenge/Solution,...

An honest SEO and discoverability review for small static or mostly-static websites, health and research sites included: a standard-library checker for titles, descriptions, canonicals, Open Graph, sitemap and robots agreement, headings, images, link text, preloads, third-party scripts and structured data, plus a workflow for reading the pages, proposing copy, getting the wording reviewed, and verifying the live site. Use when asked to improve how a site appears in search or when shared, to...

Answer a research question about cystic fibrosis, sickle cell disease, or any condition the way a researcher does, one public source at a time, and return evidence records instead of answers. Use when a task needs to search PubMed or Europe PMC, find systematic or Cochrane reviews, check whether a paper was retracted or corrected, list or inspect ClinicalTrials.gov studies, confirm a US drug approval date or read a current label section, look up or scan preprints, read a variant's ClinVar...

Senior counsel on French insurance law, grounded in primary sources (Code des assurances, Code civil, Code de la mutualité, Code de la sécurité sociale book IX, Cour de cassation case law, EU texts, ACPR doctrine). Use for French-law insurance questions: refusal of cover, nullity for false declaration, forfeiture (déchéance), exclusion clauses, two-year limitation period, proportional rule, termination and the Hamon / Chatel / Lemoine laws, motor claims and the Badinter law, dommages-ouvrage...

Ingests a scientific paper into a markdown knowledge wiki (Obsidian-compatible). Creates or updates the paper note and propagates its findings to concept and method notes, with source-fidelity checks. Use when the user says "/wiki-ingest", "add this paper to the wiki", or "ingest paper". Filesystem-first, no database or plugin required. Works for any scientific field defined in the wiki's WIKI_SCHEMA.md.