Research & Search

aibuild-lab Skill

Summarize a Slack DM thread into a 3-bullet recap.

samin12 Bundle

Research trending AI news on X/Twitter and the web, then send a daily Telegram digest of Claude-focused video ideas for Samin's YouTube channel.

manojbajaj95 Skill

Relationship-first networking skill for identifying the right people to connect with and drafting value-add outreach — not sales, not referral asks. Use when building a network for a job search, career pivot, partnership, or industry access. Covers target identification by purpose, deep LinkedIn + web research on a specific person, brainstorming 10 value-add angles, and drafting three outreach formats: cold networking email (max 200 words), LinkedIn connection request (max 300 characters), and hiring manager follow-up email (120–160 words). Never drafts generic openers or referral asks — always leads with a specific observation and genuine value.

samin12 Bundle

Use this skill when the user asks to research YouTube video ideas, find outlier videos, analyze their channel performance, find trending topics for YouTube, do TubeLab research, cross-reference outliers with channel data, or generate video idea reports.

samin12 Bundle

Coordinate multiple independent Codex tasks from one voice or chat task by creating, pinning, steering, monitoring, and synthesizing durable worker tasks. Use when the user says "read the voice orchestrator" or asks to fan out work, run several workstreams in parallel, manage Codex tasks hands-free or by voice, delegate research or implementation lanes, redirect existing tasks, or receive concise progress reports. Do not use for a single small task that is faster to complete directly.

zenx0x Skill

Compile and execute a goal, existing plan, active run, or Research Routes packet through the vNext Coordinator, Task Lanes, WorkUnits, and Workflow Packs.

zenx0x Bundle

Compare multiple research routes with explicit Claims, Evidence, unknowns, failure regimes, and reversible probes. Use when a person needs route-neutral exploration before implementation, experiment authorization, HumanDecision, or canonical-state promotion.

samin12 Bundle

Write ML papers for NeurIPS/ICML/ICLR: design→submit.

dga-sd Bundle

ค้นเอกสารกองใหญ่ในเครื่อง (เล่มมาตรฐาน, คลังความรู้, wiki/Obsidian vault, ไฟล์ .txt/.md) แล้วคืนเฉพาะ excerpt สั้น ๆ พร้อมชื่อไฟล์+เลขบรรทัดสำหรับอ้างอิง — แทนการอ่านทั้งเล่ม เข้า context ซึ่งเปลือง token มหาศาล Use this skill BEFORE answering any question whose answer should come from local documents: standards books (มสพร./มรด./TGIX), a knowledge base, a wiki, a corpus folder, or any collection of .txt/.md files — especially when the user asks "เล่มเราเขียนว่าอย่างไร", "มีนิยามคำนี้ไหม", "อ้างอิงจากเล่มไหน", or asks you to cite sources. Also use it before drafting or reviewing a document that must be consistent with existing standards, and whenever you are tempted to read a large file into context to find one fact. Do NOT use for searching code (use Grep/Glob) or for questions the local documents cannot answer.

tamvodopad Bundle

Analyze GEDCOM genealogy files to find anomalies in marriages, births, deaths. Focus on Russian Orthodox traditions (pre-1917). Use when user asks to analyze genealogy data, find atypical marriages, investigate family anomalies, validate dates, check patronymics, find brick walls (dead ends), assess data quality, find replacement names pattern, check names against Orthodox nameday calendar (svyatsy), calculate inbreeding coefficient (COI), analyze geographic migration, study longevity/lifespan statistics, analyze generation intervals, research surname etymology, analyze sibling patterns, study marriage age and endogamy, analyze multiple marriages and widowhood, study occupations and social mobility, analyze war impact on demographics, detect epidemics from death clusters, find twins, analyze orphans, study name trends, analyze godparent networks, analyze causes of death, analyze wedding witnesses and baptism sponsors, export timeline data, or get comprehensive tree statistics.

codingendless Bundle

Automatically apply only when the user asks to create, modify, fix, refactor, or otherwise write code or code-adjacent project files. Use the fastest, lowest-cost direct execution path. Do not apply to everyday non-coding work, general chat, research, planning, explanation, documents, data analysis, or read-only review.

rowland-dot Bundle

Use Xiaohongshu DianDian AI to fully summarize one saved image or video note and save the completed reply as private structured data. Use when a user asks to summarize, parse, or deeply read a single Xiaohongshu favorite through 点点 AI, especially when captions alone omit information carried by images, video, or audio.

mucjustin Bundle

Use when an MPA graduate student needs course study, case analysis, case competition entries, policy memos, literature work, research design, fieldwork, data analysis, thesis writing, defence preparation, or AI-use disclosure statements in an established MPA, public-management, or public-policy context. Generic academic, teaching, blogging, programming, data, Zotero, Obsidian, or non-MPA thesis requests do not qualify by themselves. An explicit request to convert non-MPA material into MPA work qualifies.

cuzfrog Bundle

Runs the full dev cycle (research → plan → impl → review + fix loop) given an explicit task_id.

cuzfrog Skill

Dispatches lissom-specs-reviewer and then lissom-researcher agents optionally to refine the specification and produce the research document given an explicit task_dir, which contains the original specification.

jay7793 Bundle

Run Sol-led software development in auto, solo, solo-reviewed, or team modes, with GPT-5.6 Sol owning planning, implementation or delegation, integration, verification, and delivery; terra_worker and luna_worker providing bounded implementation; bundled test-driven-development enforcing RED-GREEN-REFACTOR for production-code behavior changes; and solweaver_reviewer providing one fresh final-strict review at a declared assurance-unit boundary. Use when explicitly invoked for any software-development task, including small fixes; for software-development prompts beginning with Goal: or /goal; or when the user requests Sol-only work, a software team, agents, subagents, parallel work, or delegated implementation. Do not use for general questions, research, writing, or operations-only requests.

yanchrisdifa Skill

Benar-benar menonton video dari sebuah tautan (TikTok, YouTube, Instagram, Reels, Shorts, Twitter/X, Facebook, atau file video lokal) lalu melaporkan isinya. Pakai skill ini setiap kali user menempelkan tautan video dan bertanya "bisa nonton video ini?", "apa isinya?", "rangkum dong", "tolong cek videonya", "transkripnya apa", "watch this video", "summarize this video", atau meminta subtitle, transkrip, atau kutipan dari sebuah video. Pakai juga kalau isi video dibutuhkan sebagai bahan untuk tugas lain, misalnya menulis artikel, memverifikasi klaim, atau mengambil potongan kode yang muncul di layar.

cmoralesm Bundle

Comprehensive, extraction-verified reference for Portfolio123 (P123), the systematic equity research platform: factor investing, stock screening, ranking systems, backtesting, and the REST API with the p123api Python wrapper. Use this skill whenever the user mentions Portfolio123, P123, p123api, P123 screens or screen rules, ranking system XML, simulated strategies, buy/sell rules, or any P123 formula syntax such as Close(0), FRank, FHist, MktCap, PEExclXorTTM, ROE%TTM, SetVar, or Eval. Also use it when the user wants to write or debug P123 formulas, build or fix ranking systems, construct screens or universes, replicate academic factor strategies (value, momentum, quality, low volatility) on P123, or pull P123 data programmatically (screen_run, screen_backtest, rank_ranks, data_universe, AI Factor predictions). Covers all 4,463 factors and 465 functions of the official Factor Reference in 13 category files, plus the full REST API (39 operations) and 9 runnable example scripts.

conorpbrennan Bundle

Write an implementation plan that sub-agents can build unattended - one step per commit, one review per commit - or run a plan that already exists. Use when work is large enough to need a plan before code, when asked to plan or orchestrate a multi-step build, or when dispatching implementers and reviewers step by step. Carries the step form, the plan skeleton, the pre-run review loop and the four dispatch prompts. Not for search-shaped work (research, exploration, prototyping to find the design), where the step list cannot be fixed before dispatch.

huggingface Bundle

Papers With Code CLI (`pwc`) for searching and reading AI/ML papers, discovering recent and trending research, finding related work and paper lineage, browsing tasks, methods, conferences, organizations, frameworks, and benchmark leaderboards, and submitting authenticated paper edits through the public Papers With Code catalog. Use whenever the user asks to find papers, survey literature, compare research, inspect an arXiv paper, explore AI/ML taxonomy or conferences, discover benchmarks or state-of-the-art models, or mentions Papers With Code, `pwc`, or `pwc-cli`.

undertakingyou Skill

Draft a structured commit message from uncommitted changes (staged, unstaged, and untracked). Use when the user asks to write, draft, generate, or create a commit message, or says "what should my commit message be", "summarize my changes for a commit", etc. Do NOT use if the user asks to actually commit — this only drafts the message.

coreybrandon Skill

QRSPI Phase 0: Init — Bootstrap a feature workspace from a ticket, issue, or description. Use when the user wants to start the QRSPI workflow (Questions-Research-Spec-Plan-Implement) for a new feature, or explicitly says 'qrspi init' / 'start qrspi'.

coreybrandon Skill

QRSPI Phase 3: Design — Align on current state, desired end state, and architectural decisions. The highest-leverage review point in the workflow. Use when the user has completed qrspi-research and wants to move to design, or explicitly says 'qrspi design'.

coreybrandon Skill

QRSPI Phase 2: Research — Gather objective codebase facts to answer the questions phase, without reading the ticket. Use when the user has completed qrspi-questions and wants to move to research, or explicitly says 'qrspi research'.

coreybrandon Skill

QRSPI Phase 1: Questions — Identify knowledge gaps with targeted technical questions before research begins. Use when the user has completed qrspi-init and wants to move to the questions phase, or explicitly says 'qrspi questions'.

mine-fnl Skill

Use when designing, reviewing, or critiquing quantitative trading or investment research — alpha signals, backtests, factor models, and statistical validation. Trigger on requests like "backtest this strategy", "is this signal real?", or questions about Sharpe ratios, overfitting, look-ahead bias, or factor exposures.

yiweicreates Skill

从宽泛需求发现合适的 GitHub 开源项目，或在安全前提下把指定仓库读透。用户说“GitHub 上有没有能做 XXX 的项目 / 帮我找一批候选 / 做开源选型”时，先做需求画像、搜索矩阵、云端初筛、证据评分和选取关口；用户给项目名或 GitHub 链接时，做身份核验、作者 Brief、安全体检、隔离安装、架构与核心权衡深读、沙箱实测和证据化报告。English triggers - "find open-source projects for this need", "shortlist GitHub repos", "research this GitHub repo", "is this repo safe / worth learning from".

wjjjjjjjjjjj Bundle

对需要多来源综合与争议分析的主题开展研究，形成有证据的判断。单个事实查询、指定材料阅读和普通改稿不触发完整研究流程。

zime-skills Skill

Research a new prospect company before you reach out. Checks if they fit our ICP and who the buyers might be. Use for companies not yet in CRM. For existing accounts, use get-account instead.

antsteyer Skill

Summarize what the user has worked on since a relative date ("depuis hier", "hier midi", "cette semaine", "ce mois-ci", "depuis lundi", etc.), across the current git repo and its related sibling repo/worktrees (e.g. agorize-front + agorize-core). Reports commits grouped by ticket/branch, worktree status (clean/uncommitted/unpushed), and associated PR state. Use when the user asks "sur quoi j'ai travaillé depuis X", "qu'est-ce que j'ai fait cette semaine", "résume mon activité depuis hier", or similar.

antsteyer Skill

List all FRONT tickets for a given sprint in the Agorize PROD Jira project (agorize.atlassian.net) with content summaries and story point estimates. Use this skill when the user asks to list, show, summarize, or review front-end/FRONT tickets for a sprint, mentions a sprint number (e.g. "W17", "W18", "sprint W20"), or asks "quels tickets FRONT arrivent en W17", "résume-moi le sprint FRONT", "c'est quoi les tickets front du sprint". Always invoke this skill when a Jira sprint review for FRONT tickets is requested, even if the user just says the week number.

hiyabh Bundle

Expert skill for organizing large unstructured file collections — email exports, legal documents, NGO archives, research folders, or any mixed document archive. Use this skill whenever the user mentions having hundreds or thousands of unorganized files they need to sort, classify, or clean up. Trigger phrases include: "organize my files", "sort my archive", "classify documents", "ארגן קבצים", "מיין תיקיות", "סדר את הארכיב", "I have thousands of files", "create a folder structure", "sort my emails", "clean up my downloads", "help me organize", "file organization", "document classification", "de-duplicate files". Also trigger when the user describes a messy directory with many files and no clear structure, or when they want a systematic folder hierarchy created. This skill provides the complete methodology, Python script templates, and proven patterns for organizing any size archive safely and reproducibly.

clearailhc Bundle

【论文检索·学术 API】通过 REST API 查 PubMed/arXiv/OpenAlex/Crossref 等学术库。适用：DOI/PMID 解析、论文元数据、引用关系、开放获取链接。不适用：行业快研与商业信息（用 web_search + domain-presearch）；完整综述流程（用 literature-review）。

clearailhc Bundle

【领域·背景预研】结构化的研究领域/问题域背景快研报告（领域画像、发展脉络、方法格局、开放问题、代表工作、趋势方向）。适用：进入新领域前的快速背景调研、选题切入分析。不适用：单一流程/系统的机理深度调研（用 process-presearch）；意图模糊的早期发散（用 scientific-brainstorming）；系统文献综述（用 literature-review）。

clearailhc Bundle

【文献综述·系统检索】按 PRISMA 思路做多库检索、主题综合与引用核验。适用：研究主题的系统性文献梳理、技术路线证据链。不适用：售前企业快研（用 domain-presearch）；单篇 DOI 查找（用 paper-lookup）。

clearailhc Bundle

【引用管理·BibTeX】检索论文元数据、DOI→BibTeX、校验与去重。适用：预研/综述报告文末引用统一、交付前引用准确性检查。不适用：全文综述撰写（用 literature-review）；一般网页来源（用 web_search sources 标注）。