Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
Productivity
Design a technical event serving an in-room and a remote audience at once, after the format decision has already chosen hybrid - which sessions the remote audience gets, how far a remote attendee's voice reaches the stage, the cue track the stream needs beside the room's, and the crew split between them. Use whenever the user mentions streaming talks to a remote audience, remote Q&A, staffing the remote side, or running a day for two audiences at once - even if they never say "hybrid". Do NOT use to decide whether to run hybrid at all - use samber/dev-event-organizer-skills@event-format-selection; an event with no room is samber/dev-event-organizer-skills@virtual-event-production.
Reduce and absorb no-shows at a technical event — an overbooking ratio against the room's real capacity and a stated ceiling, the waitlist and what triggers a promotion off it, the day-of walk-in lane, and the expected-show-up number handed to catering instead of the registration count. Use whenever asked how many registrations to accept for a free event, whether to overbook, how to run an event waitlist, why half the RSVPs did not turn up, or how many people to order food for. Do NOT use for reminder email copy — use samber/dev-event-organizer-skills@event-attendee-email-sequences; the check-in desk — use samber/dev-event-organizer-skills@event-attendee-experience; or the demand verdict behind a low show-up rate — use samber/dev-event-organizer-skills@event-market-fit.
Launch the first edition of a tech event (conference, hackathon, large meetup) from zero — a minimum viable edition 1, scope discipline against the full vision, the first cross-organization team, the decision-to-doors-open reverse timeline, and a pre-mortem of first-edition-specific risks with written go/no-go dates. Use whenever asked to start a new conference or hackathon, plan a first edition, scope a minimum viable event, form a founding organizer team, or set a launch timeline for an event with no track record. Strategy layer for edition 1 only. Do NOT use to validate demand first — use samber/dev-event-organizer-skills@event-market-fit — or to grow an established event — use samber/dev-event-organizer-skills@event-growth-strategy.
Find and negotiate the venue for a technical event. Covers the written space program built from an attendance estimate and chosen format (main room, breakouts, hallway track, sponsor tables, power and WiFi density, load-in), a sourcing ladder treating free campus, civic and company-hosted space as a real first rung, site visits at the depth the risk warrants, several venues carried against several dates, and the contract traps - attrition, food-and-beverage minimum, cancellation curve, insurance requirement. Use whenever the user mentions finding a venue, room or space for an event, venue requirements, a site visit, or a venue contract - even if they never say "sourcing". Do NOT use for non-venue suppliers - use samber/dev-event-organizer-skills@event-vendor-sourcing instead.
Find and qualify speakers for a technical event outside the open call for papers - the proactive channel running alongside or instead of a CFP. Covers program gaps worth filling by invitation, sourcing channels (referral asks, expert and speaker directories, prior-talk review, community scouting), international sourcing by region and language, evidence-based Hot/Warm/Cold/Skip qualification, and a ranked shortlist carrying the "why" behind each name. Use whenever the user mentions finding speakers, a keynote or invited-speaker shortlist, sourcing international or under-represented speakers, or a programme a CFP did not fill - even if they never say "sourcing". Do NOT use to write the invitation - use samber/dev-event-organizer-skills@event-speaker-cold-outreach instead.
Approve and coordinate third-party-run satellite events around a technical event — the vetting gate, calendar deconfliction against the published grid, name and logo usage rules, and the disclaimer stating what the main event does not vouch for. Use whenever a user group, sponsor or attendee wants to run a party, meetup, dinner, workshop or afterparty during the event's dates, when someone is already using the event's name on their own page, or when fringe events collide with the main programme — even if they never say "side event." Approving nothing is often the right answer; routes trademark and liability exposure to counsel. Do NOT use for the organizer's own evening programme — use samber/dev-event-organizer-skills@event-official-social-program instead.
Turn several editions of a recurring technical event's debrief logs into operational trends and deliberate changes - a signal gate separating a one-off from a reproducible pattern, themes grouping differently-worded findings, recurrence counts across editions, a learning log persisted so it survives organizer turnover, an audit so it never grows unexamined, and a route from each resulting change to the sibling that owns it. Use whenever asked what keeps going wrong every year, how to stop relearning the same lessons, how to keep institutional memory across a team that turns over, or what several editions of evidence say to change next. Do NOT use for one edition's retrospective - use samber/dev-event-organizer-skills@event-debrief.
Recruit and organize the volunteer workforce that staffs one edition of a technical event - roster sizing built bottom-up from the posts the schedule needs standing, role definitions with real fit criteria, shift plans with arrival buffers, sign-up and confirmation, the briefing including the code-of-conduct escalation script, a named replacement protocol holding coverage, and recognition for the people who showed up. Use whenever the user mentions how many volunteers an event needs, recruiting or signing up volunteers, volunteer roles or a shift schedule, briefing volunteers, or a volunteer who does not turn up - even if they never say "volunteers". Do NOT use for volunteer care and retention - use samber/dev-event-organizer-skills@event-volunteer-experience instead.
Run a startup pitch contest or demo day inside a technical event—the pre-event application and selection funnel, the pitch format and the time it costs, the rubric dimensions an investor judge actually scores, the conflict of interest of a judge who may invest in the company just ranked, and a prize the winner can use—introductions, committed slots or equity-free cash. Use whenever the user mentions a pitch competition or demo day, opening applications for startups, picking a slate, briefing an investor jury, pitch time limits, or what the winner gets—even if they never say “pitch contest.” Routes securities and contest-law questions to counsel. Do NOT use for projects built during the event—use samber/dev-event-organizer-skills@hackathon-judging instead.
Run the speaker-side experience from the moment a speaker confirms through to the thank-you after their talk—collecting bio, headshot, A/V needs and travel details against a deadline ladder, a single named point of contact, the pre-event briefing and slide deadline, tech check and green room, documented recording consent and its per-format opt-out asymmetry, and post-event follow-up. Use whenever the user mentions what to send an accepted speaker, a green room, speaker travel or hotel, a tech check, recording consent, a speakers' dinner, or a first-time speaker's care—even if they never say “speaker experience.” Do NOT use before acceptance (samber/dev-event-organizer-skills@event-speaker-cold-outreach) or for dry runs and talk feedback (samber/dev-event-organizer-skills@event-speaker-coaching).
Query decomposition and multi-source search orchestration. Breaks natural language questions into targeted searches per source, translates queries into source-specific syntax, ranks results by relevance, and handles ambiguity and fallback strategies.
Paper workflow: read papers and produce 取景框 library cards in one go. Takes one or more arxiv links, paper URLs, PDFs, or paper names. For each paper, runs ljg-paper (generates org analysis), then ljg-library (distills the paper's 取景框 into a 2050 library card PNG). Use when the user says '论文流', 'paper flow', '读论文并做卡片', '论文卡片', or provides multiple papers and wants both analysis and cards.
Design motivation systems using Autonomy, Mastery, and Purpose (AMP) for products and teams. Use when the user mentions "intrinsic motivation", "gamification isnt working", "rewards arent working", "autonomy", "mastery", "purpose-driven", "my team is disengaged", or "how do I motivate people". Also trigger when designing onboarding progression, fixing broken gamification, or building team structures that sustain high performance. Covers why carrot-and-stick fails and how to build progress systems. For habit-forming product loops, see hooked-ux. For retention behavior design, see improve-retention.
Automatically organizes invoices and receipts for tax preparation by reading messy files, extracting key information, renaming them consistently, and sorting them into logical folders. Turns hours of manual bookkeeping into minutes of automated organization.
Generate a standup update from recent activity. Use when preparing for daily standup, summarizing yesterday's commits and PRs and ticket moves, formatting work into yesterday/today/blockers, or structuring a few rough notes into a shareable update.
Paper reverse reading: Given a paper, recursively find earlier papers it critiques and improves on (up to 5 levels), then find the latest developments after it and tell the evolution of the problem forward from its origins. Center the problem and use Feynman-style explanations to describe the problems each paper identifies and its solution innovations. Use when a user shares a paper and wants to understand its intellectual lineage, citation chain, problem evolution, or says “倒读”, “论文溯源”, “论文脉络”, “paper river”, “paper connects”, “trace back”, “这篇论文的来龙去脉”, or “论文演化”. Also trigger when the user wants to understand how a research problem evolved across multiple papers.
Word flow: deep-dive word analysis + infograph card in one go. Takes one or more English words, runs ljg-word (generates deep semantics analysis) then ljg-card -i (generates infograph PNG). Use when user says '词卡', 'word card', 'word flow', or provides English words wanting both analysis and visual card.
Implement the Entrepreneurial Operating System (EOS) to align vision and execution across a company. Use when the user mentions "EOS", "Entrepreneurial Operating System", "V/TO", "quarterly rocks", "Level 10 meetings", "accountability chart", "IDS process", "my company feels chaotic", "we keep having the same problems", or "get the whole team aligned". Also trigger when a growing company needs meeting structure, goal-setting frameworks, or a systematic way to solve recurring organizational issues. Covers the six EOS components: Vision, People, Data, Issues, Process, Traction. For team motivation design, see drive-motivation. For lean experimentation, see lean-startup.
A retrieval and Q&A assistant for local knowledge base directories. Core flow: (1) navigate the layered index (2) when encountering PDF/Excel, first read references to learn the handling method (3) retrieve after processing the file. Combine grep, Read, pdfplumber and pandas by file type for progressive retrieval to avoid loading whole files. Use when the user's question involves "answering questions/retrieving information/looking up material from a knowledge base directory".
Call the Recoup API and external connectors directly — fetch any platform resource (artists, socials, organizations, research, documents) and run connector actions (Google Docs/Sheets/Drive edits, Gmail, TikTok, Instagram). Use whenever you need raw Recoup data, a platform resource, to write curl against api.recoupable.dev, or to read/write something outside Recoup like a Google Doc URL or a spreadsheet. The plumbing every other skill rides on. To onboard or operate on an artist use the recoup-roster-* skills; for first-run connection use recoup-platform-connect-account.
Search across all connected sources in one query. Trigger with "find that doc about...", "what did we decide on...", "where was the conversation about...", or when looking for a decision, document, or discussion that could live in chat, email, cloud storage, or a project tracker.
Garbage collection for your Claude Code configuration. Periodically scans ~/.claude (skills, memory, hooks, permissions, MCP servers, caches) for redundant, stale, orphaned, or low-value items, then guides you through a confirm-each-deletion cleanup. Use when you say “clean up my config,” “config GC,” “too many skills,” “audit my setup,” “my .claude is bloated,” or request a periodic configuration review.
Sync tasks and refresh memory from your current activity. Use when pulling new assignments from your project tracker into TASKS.md, triaging stale or overdue tasks, filling memory gaps for unknown people or projects, or running a comprehensive scan to catch todos buried in chat and email.
Lark video meetings: query meeting records and artifacts (minutes/transcripts/Miaoji), search/upload/download/edit Miaoji, and have bots join meetings; query ongoing meetings and real-time meeting content (speaking/chat/shared documents), answer questions about meetings, and send in-meeting chat messages/reactions. Look up related information using meeting_id, meeting_no, event_id, note_id, minute_token, vc-node-id, or a Miaoji URL. Use lark-calendar for scheduling, availability, and room management.
Send an email via the Recoup API reliably—a Node helper that serializes the body correctly and fails loudly instead of sending an empty email. ALWAYS use this script to send email whenever a task needs to email a report, summary, or notification to the account owner or a recipient. Never hand-build `curl … -d "{…}"` with inline `jq`/shell interpolation; it silently produces emails with an empty body and only the “Message from Recoup” footer. Pairs with recoup-platform-api-access, which it sends through.
Use when scoping a competitive landscape — identifying, categorising, and score-filtering a competitor set before any benchmarking begins. Decides who counts as a competitor, which tier they belong to, and which sources to mine. First step in the three-skill competitive pipeline; precedes benchmark-methodology.
Deep research on any topic — broad parallel web searches, multi-source validation, confidence tracking, and a cited Markdown report. Use whenever the deliverable is a thorough sourced report rather than a quick answer: 'research <topic>', 'deep dive on X', 'analyze the landscape', 'competitive analysis', 'compare these options', 'who are the players in Z', 'literature review', 'background on Y', 'what papers exist on X', 'product teardown', 'regulatory overview', 'funding landscape', 'what trends are emerging in X', 'patent landscape', 'community health', or any request requiring scanning many sources and producing a cited written analysis. Covers 11 research types: market (TAM/SAM, segments, pricing, trends), domain (industry structure, ecosystem, regulatory overview), technical (architecture, tooling, benchmarks, technology evaluation), competitive (competitor teardown, positioning, win/loss), product (feature analysis, reviews, teardowns, roadmap signals), academic (literature survey, citation networks, key authors), person/org (due diligence on a company or public figure), financial (funding landscape, valuation multiples, revenue signals), legal (IP, patent landscape, litigation, compliance), trend (emerging signals, foresight, scenario mapping), community (ecosystem health, key voices, governance, fragmentation). Trigger even when phrased casually: 'look into X', 'what's the deal with Y', 'dig into Z', 'I need to understand the space', 'catch me up on X'. Do NOT use for single-fact lookups or one-off web questions.
Initialize the productivity system and open the dashboard. Use when setting up the plugin for the first time, bootstrapping working memory from your existing task list, or decoding the shorthand (nicknames, acronyms, project codenames) you use in your todos.
Assemble scored competitor profile cards (from benchmark-methodology) into a decision-grade competitive report with landscape map, competitor tiers, benchmarking matrix, white-space analysis, strategic recommendations, and team alignment trigger questions. Use when presenting competitive findings to leadership or a board, writing a competitive landscape report, or as the final step of the competitive analysis pipeline.
Build a custom cmux sidebar from a plain-language request. Use when the user asks for a custom sidebar, a sidebar that shows their workspaces/tabs/PRs/clock, a vibe-coded sidebar, or anything involving files in ~/.config/cmux/sidebars/. Covers authoring the interpreted SwiftUI-style file, enabling the beta flag, selecting it, and iterating with hot reload
Information question-extraction machine. Given an article, paper, or book, extract its core ideas as Q-A pairs — the Question cuts to the vital point, not textbook-like; the Answer is concise and clear, with a formal closure and a complete logical chain. Readers follow the chain of Q, and each A drives in a nail, reproducing the author's entire reasoning. Use when the user says "Q&A", "QA", "ask questions", "extract questions", "/ljg-qa", or shares an article/paper/book and asks for Q-A extraction. Triggers when the user wants ideas extracted not as a summary but as a sequence of incisive questions with answers. NOT for FAQ generation, glossary creation, or comprehension quizzes — this is intellectual scaffolding, not study aids.
Scaffold a music company's working OS in its own repository, with a self-managing workspace, read-only health checks, reusable knowledge and portable skills. Use for "set up our label OS", "build our publishing workspace", "organize our catalog operations repo", or "mirror my roster into folders". Select the actual core objects (artists/releases, compositions/parties or catalog cases); use supported Recoup records when available and private file workflows otherwise. No artist roster, valuation or account claim is required for every music business.
Add or attach an artist to your roster and enrich them — verify Spotify identity, resolve the artist in the requested workspace, then research, catalog, socials and knowledge base. Supports deliberate manual onboarding when Spotify identity is unavailable. Use for "create/add/onboard an artist", "set up [artist]", "we just signed [artist]", or "new signing". To see your roster use recoup-roster-list-artists; to work inside an existing artist's files use recoup-roster-manage-artist.
Work inside one artist's folder — read and update their context, brand, songs, and releases. Use for "organize [artist]'s files", "update [artist]'s brand/context", "what's in [artist]'s workspace", or any task about a named artist's files. To list your whole roster use recoup-roster-list-artists; for research/metrics on an artist use recoup-research-artist-overview.
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.