Top Agent Skills

Query and analyze SEC filings using EdgarTools

Econometrics skill for generating publication-quality figures for top economics journals. Activates when the user asks about: "econometric figure", "publication figure", "journal figure", "AER figure", "QJE figure", "event study plot", "coefficient plot", "coefplot", "binned scatter", "binscatter", "RDD plot", "parallel trends plot", "kernel density", "distribution plot", "time series plot", "map", "figure formatting", "academic plot", "论文图表", "学术图", "系数图", "事件研究图", "散点图", "分布图", "趋势图", "回归可视化"

Generate pixel art SVG illustrations for READMEs, docs, or slides. Use when user says "画像素图", "pixel art", "make an SVG illustration", "README hero image", or wants a cute visual.

Document datasets, variables, sources, and merge keys for replication

This skill should be used when the user asks to "/do-agent", asks for "Execute complex tasks using multi-agent architecture with context protection", or needs the workflow previously provided by the /do-agent slash command.

Summarize academic papers, extract key findings, and identify research gaps

Checklist of empirical robustness tests for finance/economics papers

Generate and rank research ideas given a broad direction. Use when user says "找idea", "brainstorm ideas", "generate research ideas", "what can we work on", or wants to explore a research area for publishable directions.

Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script. Use when user says "做PPT", "做幻灯片", "make slides", "conference talk", "presentation slides", "生成slides", "写演讲稿", or wants beamer slides for a conference talk.

Writes rigorous mathematical proofs for ML/AI theory. Use when asked to prove a theorem, lemma, proposition, or corollary, fill in missing proof steps, formalize a proof sketch, 补全证明, 写证明, 证明某个命题, or determine whether a claimed proof can actually be completed under the stated assumptions.

Search and analyze research papers, find related work, summarize key ideas. Use when user says "find papers", "related work", "literature review", "what does this paper say", or needs to understand academic papers.

Verify research idea novelty against recent literature. Use when user says "查新", "novelty check", "有没有人做过", "check novelty", or wants to verify a research idea is novel before implementing.

Compile LaTeX paper to PDF, fix errors, and verify output. Use when user says "编译论文", "compile paper", "build PDF", "生成PDF", or wants to compile LaTeX into a submission-ready PDF.

Workflow 3: Full paper writing pipeline. Orchestrates paper-plan → paper-figure → paper-write → paper-compile → auto-paper-improvement-loop to go from a narrative report to a polished, submission-ready PDF. Use when user says "写论文全流程", "write paper pipeline", "从报告到PDF", "paper writing", or wants the complete paper generation workflow.

Correct grammar errors, typos, and improve academic readability in LaTeX, Markdown, or plain-text manuscripts. Scans the document once, then walks through every issue one-by-one asking for approval before applying each fix. Trigger when the user says "readability", "check grammar", "fix typos", "proofread for grammar", "improve readability", "polish wording", "语言润色", "修语法", or asks you to clean up the prose in a paper/chapter/section without wanting full content restructuring.

Draft a structured grant proposal from research ideas and literature. Supports KAKENHI (Japan), NSF (US), NSFC (China, including 面上/青年/优青/杰青/海外优青/重点), ERC (EU), DFG (Germany), SNSF (Switzerland), ARC (Australia), NWO (Netherlands), and generic formats. Use when user says "write grant", "grant proposal", "申請書", "write KAKENHI", "科研費", "基金申请", "写基金", "NSF proposal", or wants to turn research ideas into a funding application.

Workflow 1: Full idea discovery pipeline. Orchestrates research-lit → idea-creator → novelty-check → research-review to go from a broad research direction to validated, pilot-tested ideas. Use when user says "找idea全流程", "idea discovery pipeline", "从零开始找方向", or wants the complete idea exploration workflow.

Deploy and run ML experiments on local or remote GPU servers. Use when user says "run experiment", "deploy to server", "跑实验", or needs to launch training jobs.

Fetch economic data from FRED, World Bank, BLS, OECD, and Yahoo Finance

Searches multiple web sources, synthesizes findings, and produces cited research reports using delegated subagents. Use when the user asks to research a topic online, search the web, look something up, find current information, compare options, or produce a research report.

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 communication-system research and the user wants papers, related work, a survey, or a landscape summary. Search Zotero, Obsidian, and local paper folders first when available, then search IEEE Xplore, ScienceDirect, ACM Digital Library, and broader web in that order.

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, simplicity, and any LLM / VLM / Diffusion / RL-based contribution.

Generate Mermaid diagrams from user requirements. Saves .mmd and .md files to figures/ directory with syntax verification. Supports flowcharts, sequence diagrams, class diagrams, ER diagrams, Gantt charts, and 18 more diagram types.

Use when experiments complete to judge what claims the results support, what they don't, and what evidence is still missing. Codex MCP evaluates results against intended claims and routes to next action (pivot, supplement, or confirm). Use after experiments finish — before writing the paper or running ablations.

Use when starting a new research project, exploring a research idea, deciding whether a question is viable, or before touching code or data for a new paper. Runs a research-focused brainstorm that clarifies research question, identification strategy, data feasibility, and contribution before any implementation.

Clean and transform messy data for analysis in Python, R, or Stata

Use when main results pass result-to-claim (claim_supported=yes or partial) and ablation studies are needed for paper submission. Codex designs ablations from a reviewer's perspective, CC reviews feasibility and implements.

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.

Search published venue papers (IEEE, ACM, Springer, etc.) via Semantic Scholar API. Complements /arxiv (preprints) with citation counts, venue metadata, and TLDR. Use when user says "search semantic scholar", "find IEEE papers", "find journal papers", "venue papers", "citation search", or wants published literature beyond arXiv preprints.

Comprehensive exploratory data analysis with publication-quality descriptive tables, correlation matrices, distribution plots, and assumption testing. Generates a standalone EDA report with Table 1 (gtsummary/great_tables), correlation heatmap, distribution diagnostics, VIF for multicollinearity, and normality/homoscedasticity tests. All figures are APA-formatted and colorblind-safe. Use when the user says "exploratory analysis," "EDA," "descriptive statistics," "explore the data," "Table 1," "correlations," "distributions," or when /data-clean completes successfully. Triggers on "EDA," "descriptive," "Table 1," "explore," "correlations."

Causal inference framework for answering "does X cause Y?" beyond correlation. DoWhy (Microsoft Research) provides the identify-estimate-refute loop: define a causal graph (DAG), identify the causal effect using backdoor/frontdoor/instrumental variable criteria, estimate treatment effects with multiple estimators, and validate results with automated refutation tests. Use when: distinguishing causation from correlation, estimating treatment effects (ATE, ATT, CATE), designing and analyzing A/B tests with confounders, using instrumental variables, performing counterfactual reasoning ("what would have happened if..."), validating causal claims with sensitivity analysis, working with observational data where randomization is impossible, or any analysis where the question is "what is the CAUSAL effect of X on Y" rather than just "how do X and Y relate?"

Orchestrate the complete post-first-draft polishing pipeline for an academic LaTeX paper by invoking five existing skills in fixed order: (1) paper-polish, (2) paper-self-revise, (3) paper-style, (4) paper-polish again, (5) reference-verify. Trigger when user says "paper pipeline" / "paper-pipeline" / "论文流水线" / "全流程打磨" / "一条龙打磨" / "初稿打磨" / "full polish pipeline" / "run the whole pipeline", or wants the entire post-draft polishing sequence run on a paper folder. Use this skill whenever the user asks for several paper-finishing steps (polish + revise + style + reference check) on one manuscript in one go, even if they don't name every individual skill.

Full autonomous research workflow — brainstorm, plan, implement, review, and document

Full research pipeline: Workflow 1 (idea discovery) → implementation → Workflow 2 (auto review loop). Goes from a broad research direction all the way to a submission-ready paper. Use when user says "全流程", "full pipeline", "从找idea到投稿", "end-to-end research", or wants the complete autonomous research lifecycle.

Use when a research implementation plan exists in docs/superpapers/plans/ and the user is ready to execute it — collecting data, running analysis, producing outputs, writing the paper. Orchestrates task execution with replication-driven verification and two-stage review at phase boundaries.

Systematic audit and review by Referee 2. Two modes — "deck" reviews slide presentations for rhetoric, visual quality, and compile cleanliness; "code" performs cross-language replication and econometric audit of empirical pipelines. Use when reviewing slides, auditing code, or verifying replication.

APSA style checker for Quarto (.qmd) files. Checks numbers, capitalization, abbreviations, italics, in-text citations, titles of works, neutral and unbiased language, and APSA-specific terminology against the APSA Style Manual for Political Science (2018, updated 2023). Produces a structured markdown report organized by document section — never modifies the source file. Use when asked to check APSA style, fix citations, review capitalization, check number formatting, or flag biased language in a .qmd document. For grammar, spelling, and punctuation, use the proofread skill instead. Supports an optional output-file argument and an optional @sec-label argument to restrict checking to one section.

Prose quality checker for Quarto (.qmd) files, grounded in William Zinsser's *On Writing Well* (30th Anniversary Edition). Checks for clutter, weak verbs, hollow qualifiers, clichés, inflated academic voice, poor leads and endings, pronoun and tense inconsistency, and unclear explanation. Produces a structured markdown report organized by document section — never modifies the source file. Use when asked to improve prose quality, tighten writing, reduce clutter, or apply Zinsser's writing principles to a draft. For grammar and punctuation, use the proofread skill. For APSA style rules, use the apsa-style skill. Supports an optional output-file argument and an optional @sec-label argument to restrict checking to one section.

Monitor running experiments, check progress, collect results. Use when user says "check results", "is it done", "monitor", or wants experiment output.

Patterns for Bayesian inference in R using brms, including multilevel models, DAG validation, and marginal effects. Use when performing Bayesian analysis.

Activate when the user is drafting any academic economics content from scratch (e.g., outlines, abstracts, introductions, data/methods/identification sections, results narratives, conclusions, referee responses, table/figure captions) and needs economics-specific structure, phrasing options, and quality checks to produce publication-ready prose.

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page.

Design and format publication-quality tables: column order, row grouping, notes, precision, reproducibility.

Find free legal full-text versions of scholarly articles via Unpaywall

Scaffold a new research project with standard directory structure, CLAUDE.md template, and documented README. Use this at the start of every new project to ensure consistent organization.

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".

所有联网操作必须通过此 skill 处理，包括：搜索、网页抓取、登录后操作、网络交互等。 触发场景：用户要求搜索信息、查看网页内容、访问需要登录的网站、操作网页界面、抓取社交媒体内容（小红书、微博、推特等）、读取动态渲染页面、以及任何需要真实浏览器环境的网络任务。

Expert jq usage for JSON querying, filtering, transformation, and pipeline integration. Practical patterns for real shell workflows.

Dynamic multi-agent workflows — plan first, then orchestrate parallel agents with adversarial verification via the local odw daemon. Use when the user asks for a "workflow", says "ultracode", or hands you a task spanning many files/items that benefits from parallel agents.

Fetch any X/Twitter post as clean LLM-friendly JSON. Converts x.com, twitter.com, or adhx.com links into structured data with full article content, author info, and engagement metrics. No scraping or browser required.

Cirq is Google Quantum AI's open-source framework for designing, simulating, and running quantum circuits on quantum computers and simulators.

Control cmux workspaces, panes, surfaces, and agent sessions safely from macOS terminal workflows.

Anti-over-engineering guardrail that activates when an AI coding agent expands scope, adds abstractions, or changes files the user did not request.

Runs repeatable AI work as checked, budgeted workflow files.

Build content-focused websites with Astro — zero JS by default, islands architecture, multi-framework components, and Markdown/MDX support.

Write efficient C code with proper memory management, pointer

Help a user learn a topic through adaptive tutoring, lesson planning, practice, retrieval checks, explanations, study guides, or exercises. Use when the user asks to learn, understand, practice, drill, review, study, or be tutored on something.

Launch RSVP speed reader for text

SymPy is a Python library for symbolic mathematics that enables exact computation using mathematical symbols rather than numerical approximations.

ALWAYS use this skill when committing code changes — never commit directly without it. Creates commits following Sentry conventions with proper conventional commit format and issue references. Trigger on any commit, git commit, save changes, or commit message task.