Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Productivity
Comprehensive audit of a paper's citations. Verifies every reference exists, checks claim-source alignment, identifies missing seminal works, and analyzes citation patterns for bias. Use when asked to check references, verify citations, or audit a bibliography.
CNKI advanced search automation tool for papers. When the user provides research keywords or a research topic, it automatically runs a three-stage search: ① combined subject search (reverse onion-peeling method, combining multiple groups of core concepts with AND to get directly relevant literature); ② independent supplementary search (searching each core concept group separately to get indirectly relevant background and theoretical literature); ③ aggregation and merging (after deduplication, displaying by category in color and outputting a single Excel file). Triggers: the user mentions needing to search papers on CNKI, advanced search, searching CSSCI/C-journal papers by keyword, downloading bibliographic records, getting paper abstracts, or sorting results by citation count; or says "help me search CNKI for papers related to XX", "use CNKI advanced search to find C-journal papers on XX topic", "help me search CSSCI papers for the keyword XX".
Interview coaching system. Activates for any coaching command: prep, round, sync, strategy, stories, mock, outreach, thankyou, analyze, progress, practice, research, decode, fit, apply, feedback, reflect, salary, negotiate, hype, pitch, resume, linkedin, present.
Conduct deep adaptive interviews to extract and organize knowledge into structured files and folders. Use when user says "interview me about X", "deep interview", "extract my knowledge", "build knowledge base from interview", "ask me questions about X", or wants to systematically capture expertise on any topic through guided questioning.
Academic research experiment design skill — a complete process from research hypothesis to a reproducible experiment plan. Use this skill whenever the user needs to plan experiments, design an ablation study, choose baselines, decide on evaluation metrics, or asks "which experiments should I run". Trigger words include: experiment design, ablation, baseline, what experiments to run, evaluation metric, how to validate a method. Applies to experiment planning in fields such as machine learning, NLP, and CV.
Workspaces for agentic teams. Complete agent guide with all 19 consolidated tools using action-based routing — parameters, workflows, ID formats, and constraints. Use this skill when agents need shared workspaces to collaborate with other agents and humans, create branded shares (Send/Receive/Exchange), or query documents using built-in AI. Supports ownership transfer to humans, workspace management, workflow primitives (tasks, worklogs, approvals, todos), and real-time collaboration. Free agent plan with 50 GB storage and 5,000 monthly credits.
Feishu voice message sender. Sends MP3 audio files as voice clips to Feishu direct or group chats (not as file attachments). Use when: sending voice messages, voice clips in Feishu, converting MP3 to Feishu voice, broadcasting voice messages, voice notifications, private voice messages, automated voice push. Related: feishu-video-sender, feishu-group-welcome. Part of the Feishu automation toolkit by UniqueClub.
Intelligently organizes your files and folders across your computer by understanding context, finding duplicates, suggesting better structures, and automating cleanup tasks. Reduces cognitive load and keeps your digital workspace tidy without manual effort.
Create and refine HEARTBEAT.md files for murmur — a CLI daemon that runs scheduled Claude prompts on a cron or interval schedule. Use this skill when the user wants to set up a recurring automated action (e.g., "monitor my GitHub issues", "check Hacker News for AI articles", "watch my endpoints", "send me a daily digest"). Guides the user through an interview, drafts the heartbeat prompt, tests it, and registers it with murmur's scheduler. Triggers: heartbeat, murmur, recurring task, scheduled action, cron, monitor, watch, automate, periodic check, scheduled prompt.
Query and analyze scholarly literature using the OpenAlex database. This skill should be used when searching for academic papers, analyzing research trends, finding works by authors or institutions, tracking citations, discovering open access publications, or conducting bibliometric analysis across 240M+ scholarly works. Use for literature searches, research output analysis, citation analysis, and academic database queries.
Academic paper review skill — perform an in-depth paper review through a structured four-step process, covering critical review, score prediction, key point refinement, and formal review output. Use this skill whenever the user needs to review a paper, simulate a reviewer's reaction, assess whether a paper can be accepted, or help judge a paper's strengths and weaknesses. Trigger words include: review this, review my paper, what would reviewers say, can this get in, paper review, give a score, find weaknesses. Applies to review simulation and evaluation of any academic paper.
Assist in writing, structuring, and polishing personal statements and self-introductions. Use when the user mentions writing a PS, personal statement, self-introduction, application essay, motivation letter, statement of purpose, cover letter, or personal bio — for study abroad applications, graduate/PhD admissions, scholarship applications, job applications, or any similar context.
Plan a round of user interviews conducted by PostHog''s AI voice agent (a "robo interviewer") — the automated voice-agent interview product. Captures a UserInterviewTopic (who to target, what to ask, framing context, question list) and calls user-interview-topics-create. ONLY trigger when the user clearly wants an AI voice agent to actually run the interview calls (e.g. "set up robo user interviews", "have the voice agent interview these users"). Do NOT trigger for ordinary user research that does not involve the voice agent — finding or shortlisting users to talk to ("who''d be a good fit to interview about Y"), planning questions for a human-run interview, or analysing feedback are audience discovery, handled with normal data queries, not this skill. Also do NOT trigger for uploading a recorded interview audio file or browsing topics with user-interview-topics-list. When intent is ambiguous, first confirm what kind of research it is and whether they want an AI voice agent to conduct it (see Step 0).
Release planning and roadmap design skill. Starting from quarterly goals, team capacity and dependency information, it outputs an executable release roadmap (milestones, dependency risks, buffer strategy, success metrics per phase). Trigger conditions: the user mentions "roadmap", "release planning", "milestones", "quarterly plan", "iteration planning", "release scheduling", "project scheduling", "dependency mapping", "risk contingency", "buffer plan", "iteration plan", "sprint planning", "release plan", "OKR breakdown" and similar keywords. Also applies when: the user provides quarterly goals/team capacity/dependency information and asks for a schedule; the user asks to break strategic goals into executable milestones; the user asks to assess project risks and buffer plans; the user asks to create a cross-team release plan. Typical input: quarterly goals + team capacity + dependency information + constraints. Not applicable to: daily standups (direct communication), single requirement design (use pm-prd-writer), data analysis (use pm-analytics).
Evaluating scientific evidence and claims. Covers study design hierarchy (RCT to expert opinion), effect sizes (OR, RR, NNT, Cohen's d), confounding, p-value vs clinical significance, GRADE quality assessment, reproducibility, and bias types (selection, information, confounding, reporting). Use when reading a paper or assessing claims.
Evaluate scientific claims and evidence quality. Use for assessing experimental design validity, identifying biases and confounders, applying evidence grading frameworks (GRADE, Cochrane Risk of Bias), or teaching critical analysis. Best for understanding evidence quality, identifying flaws. For formal peer review writing use peer-review.
Synthesis judgment for deep research; use when multiple sub_reports are complete and you need to produce synthesis.md that clarifies the main thesis, strength of evidence, cross-dimension consensus, key conflicts, explanations for conflicts, uncertainties, and the answer to the original question. It is for "thinking the conclusion through first," not for writing the final report directly.
Plan, run and report systematic reviews and meta-analyses to PRISMA 2020 standards. Use when writing a review protocol (PROSPERO/OSF), building reproducible database search strings (PubMed, Embase, Scopus, Web of Science), deduplicating exports, organizing title/abstract and full-text screening, assessing risk of bias, drawing the PRISMA flow diagram, or completing the PRISMA 2020 checklist. Includes a dedup tool and a checked PRISMA diagram generator.
Unified orchestration of keyword-driven searches across public web pages, latest AI developments, vertical domains, and social platforms, returning standardized candidates; supports bounded X Research, explicit site Map, and lightweight verification of candidates obtained in this search. Not for reading or downloading known URLs provided directly by the user; use yichen-content-archive for such tasks.
Yichen's personal general entry point for internet research. Use for cross-platform, multi-stage research where the user has not chosen a tool, or explicitly requests systematic, in-depth research on a company, product, person, technology, industry or domain with longitudinal and cross-sectional analysis, development history plus current-state comparison, and source constraints; first generate a plan with a deadline and evidence gates, then route search discovery, candidate verification, limited archiving, on-demand transcription and evidence synthesis to yichen-unified-search, yichen-content-archive, yichen-bookmarks-export and yichen-asr. If the user only wants search, only handling known links, only exporting personal bookmarks, or only transcribing existing audio/video, use the corresponding child Skill directly. Use when an internet-research request spans multiple stages, requires evidence-grounded longitudinal and cross-sectional synthesis, the correct child route is unclear, or the user explicitly invokes $yichen-web-research.
Complete academic research skill suite covering the full pipeline: paper reading (read/explain papers with storytelling), idea generation (brainstorm research directions), experiment design (plan experiments, ablation, baselines), proof writing (mathematical proofs, LaTeX theorems), paper writing (draft to camera-ready for top venues like NeurIPS/ICLR/ACL), paper review (structured 4-step review with scoring), and professor fit analysis (evaluate advisors, cold emails, interview strategy). Use when the user wants the full research workflow or is unsure which specialized skill to pick. Trigger keywords: academic research suite, research pipeline, read paper, brainstorm, experiment design, prove, write paper, review, professor fit, advisor, cold email, LaTeX, research, NeurIPS, ICLR, ACL, arXiv, 讀論文, 寫論文, 審稿, 實驗設計, 數學證明, 研究方向, 教授分析, 選指導教授.
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.
Guidance for open science practices — preregistration (OSF Registries, AsPredicted, PROSPERO, ClinicalTrials.gov), open data and FAIR principles, repository choice (Zenodo, Dryad, Figshare), open access routes (Green/Gold/Diamond), Creative Commons licensing, reproducible computational workflows (Docker, Binder, Code Ocean), registered reports, open peer review, and TOP Guidelines. Use when preregistering a study, writing the FAIR data-sharing and repository section of a grant data management plan (NSF, NIH, ERC, UKRI), choosing a data repository, navigating open access, or building reproducible analyses. For human-subjects ethics, IRB applications, informed consent, or GDPR/HIPAA compliance, defer to alterlab-research-ethics. Part of the AlterLab Academic Skills suite.
Write LaTeX ML/AI review articles for arXiv using the IEEEtran template and verified BibTeX citations. Use when writing, planning, or continuing an arXiv review/survey paper, or when validating and repairing citations in an existing LaTeX project.
Given a concept, mine its neighbouring concepts from abstract texts retrieved from real search results, and draw a knowledge map centred on that concept — every node anchored to real literature, with no nodes added from memory. Also output a Mermaid concept diagram (renderable in Claude Code / GitHub / Obsidian) and a standalone screenshot-able HTML map, plus 3–6 recommended introductory papers. **Trigger this skill proactively in the following cases**: the user says "draw me a knowledge map of this concept", "what related concepts surround this term", "map out the concept network in this direction", "give me a concept diagram", "what are the core concepts in this field", "knowledge map", "concept map", "map out this concept", "draw a knowledge map of X", "what concepts surround X", "show me the concept network around this term", "concept map for X" — even if the user does not say "bear-map", use this skill whenever the intent is **to understand the knowledge landscape around a single concept**. Not applicable to: finding supporting literature for a claim (use bear-support), finding opposing literature (use bear-counter), checking topic novelty (use bear-scoop), or tracing the history of evolution (use bear-trace).
Given a concept or field, draw both its conceptual landscape (space: what this field looks like today) and its line of evolution (time: how this field got to where it is), helping you quickly build a three-dimensional understanding of an unfamiliar field in a single comprehensive report. Built on real search (scimaster-cli), never fabricating citations. **Trigger this skill proactively in the following cases**: the user wants to "quickly get started in a field", "get a handle on this direction", "the core concepts and developmental history of this field", "help me build a cognitive framework for this direction" — even if the user does not say "bear-onboard", use this skill whenever the intent is **to understand both the conceptual landscape and the line of evolution of an unfamiliar field**. Not applicable to: drawing only a concept map (use bear-map), tracing only the origins (use bear-trace), finding literature for a claim (use bear-support / bear-counter), or checking topic novelty (use bear-scoop).
Given a topic or a seed paper, dig forward into which predecessor works it builds on (tracing back layer by layer), then add the latest progress going forward, explaining clearly how this problem evolved from its origins to the present. Built on real search (scimaster-cli), with every node anchored to a paper actually retrieved. **Trigger this skill proactively in the following cases**: the user asks "where did this problem come from", "help me trace this direction", "how did this field develop", "whose work is this paper built on", "tell me the history of this direction from the start", "I am new to a field and want to understand its line of evolution first" — even if the user does not say "bear-trace", use this skill whenever the intent is **to understand the ins and outs of a research problem or field**. Add `-d N` to control the number of layers traced back (default 3 layers). Not applicable to: finding supporting literature for a claim (use bear-support), finding opposing literature (use bear-counter), checking topic novelty (use bear-scoop), or concept maps (use bear-map).
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.