Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
Build quick IRR/MOIC sensitivity tables for PE deal evaluation. Models returns across entry multiple, leverage, exit multiple, growth, and hold period scenarios. Use when sizing up a deal, stress-testing assumptions, or preparing IC returns exhibits. Triggers on "returns analysis", "IRR sensitivity", "MOIC table", "what's the return at", "model the returns", or "back of the envelope"
Use when the user asks to “QA my conversion tracking before launch”, “check my UTMs / pixel / event firing”, “set up a tracking pre-flight”, or “set the dedup rule so Meta and Google stop double-counting”; builds and fixes the measurement plumbing — conversion-event firing, UTM hygiene, cross-platform dedup rules, attribution-window alignment, and offline/iOS-ATT modeled-gap flags — as a pre-flight checklist plus a UTM/event-spec builder. Not for scoring R1/R2 — that is a scored veto in ad-account-auditor; not for account structure — use campaign-architect. 付费广告转化追踪QA/UTM规范/跨平台去重
A McKinsey-style problem-solving system. Starting from business problems, it uses hypothesis-driven structured analysis to generate McKinsey-style research reports and PPTs. Combines the Problem Solving methodology, MECE principle, Issue Tree breakdown, Hypotheses formation, Dummy Page design, intelligent data collection, and professional PPT generation.
Use when platform-reported conversions disagree with GA4/ecommerce, when you suspect Meta and Google are double-counting the same sales, or for a standing (monthly) reconciliation workbook that de-dups stacked credit against an order-ID truth set, normalizes attribution windows and currency, compares attribution models, and reads incrementality from a geo/holdout test. Not for the point-in-time R2 veto or RQS gate — use ad-account-auditor; not for the ROI/ROAS ratio math itself — use roi-calculator; not for organic dark-social share attribution or GA4 direct-traffic decomposition — use dark-social-attributor. 付费广告归因对账/去重/增量
Prepare for client review meetings with portfolio performance summaries, allocation analysis, talking points, and action items. Pulls account data together into a concise, meeting-ready format. Use before quarterly reviews, annual checkups, or ad hoc client meetings. Triggers on “client review,” “meeting prep for [client],” “quarterly review,” “prep for [client name],” or “client meeting”
Use when the user asks to “mine my search terms”, “find new keywords from converting queries”, “build a negative-keyword list”, or “cut wasted paid spend”; harvests converting queries into new keywords/ad groups, builds a standing negative-keyword list and an n-gram waste report from the search-terms export, and delivers a maintenance diff (add / negate / move). Not for account structure — use campaign-architect; not for budget split — use budget-optimizer; not for computing the final RQS — use ad-account-auditor. 付费广告搜索词挖掘/否定关键词/浪费词清单
Make Chalk Python resolvers statically accelerable. Use when the user asks to accelerate resolvers, fix static-acceleration diagnostics, check why a resolver "can't be accelerated", or speed up Python feature pipelines in a Chalk project (a repo with chalk.yaml/chalk.yml).
Use when the user asks to "build audience segments from my customer list", "make value-based / lookalike seed lists", "set up exclusion / suppression segments", or "map audiences to funnel stages across platforms"; turns the user's OWN customer/CRM/GA4 export into seed audiences, value-based lookalike SEED lists, exclusion/suppression segments, and a cross-platform funnel-stage targeting map, informing the ROAS A (Audience) dimension. Not for building account structure or match types — use campaign-architect; not for organic SERP intent — use keyword-research. Paid-ad audience segmentation / seed audiences / exclusion audiences / lookalike seed audiences
Find and debug Chalk query plan cache misses from engine logs. Use when a query is replanning on every request, an “ad-hoc query request missed the plan cache” warning appears, latency spikes from repeated planning, or someone asks why two seemingly identical queries don’t share a cached plan. Drives the `chalk logs` CLI search
Use when auditing a paid ad account for incremental contribution, wasted spend, or measurement integrity before scaling; runs a typed 20-item ROAS profile with verified vetoes and a SHIP/FIX/BLOCK/UNDECIDED gate on own exported data. Not for campaign structure design — use campaign-architect; not for creative production — use ad-creative-builder. 付费广告账户审计/ROAS评分
Use when the user asks to "design an A/B test", "set up a creative/landing test", "run an incrementality test", or "is this result statistically and practically material?"; produces a hypothesis, variant matrix, sample-size/duration/power plan, and a documented effect/uncertainty read from own exported results. It applies only a precommitted owner-approved action rule; the statistical helper never chooses a business action. Not for producing variants — use ad-creative-builder; not for reading back one shipped change — use paid-measurement-loop. 广告AB测试设计/实验设计/显著性判定/增效测试
Use when the user asks to "analyze influencer campaign performance", "compare influencers", or "find what content worked"; produces metric scorecards vs target and benchmark, platform/influencer/content rankings, engagement-quality and sentiment reads, conversion-attribution breakdowns, and ranked learnings. Not for dollar-level return math — use roi-calculator. 达人营销效果分析/投放复盘
Use when the user asks to “check pacing,” “am I over/under-spending,” “is this campaign on track to hit budget,” or “why did spend spike/stall mid-flight”; returns a spend-vs-target-curve read, learning-phase status, an over/under-delivery call, and a reallocation trigger. Not for initial budget allocation — use budget-optimizer; not for choosing the bid strategy — use bid-strategy-planner; not for the RQS gate — use ad-account-auditor. Paid ad budget pacing monitor / overspending or underspending / in-flight pacing
Interactive scientific and statistical data visualization library for Python. Use to create charts and visualizations including scatter plots, line charts, bar charts, heatmaps, 3D plots, geographic maps, statistical distributions, financial charts, and dashboards. Supports both quick visualizations (Plotly Express) and fine-grained customization (graph objects). Outputs interactive HTML or static images (PNG, PDF, SVG)
Use when the user asks to "optimize my Shopping feed", "fix product disapprovals", "improve product titles/attributes", or "build feed-driven PMax asset groups"; audits and rewrites the Shopping/Performance Max product feed — title/description patterns, required and recommended attributes, GTIN/availability/price hygiene, disapproval triage, and feed-driven asset-group / listing-group structure — informing the ROAS O (Offer) dimension. Not for text ad copy — use ad-creative-builder; not for scoring the account or the RQS — use ad-account-auditor. 商品Feed优化/购物广告Feed/商品标题优化/商品禁投修复
Writes and executes SQL queries against the data warehouse using dbt's Semantic Layer or ad hoc SQL to answer business questions. Use when a user asks about analytics, metrics, KPIs, or data (e.g., “What were total sales last quarter?”, “Show me top customers by revenue”). NOT for validating, testing, or building dbt models during development.
Update financial models with new data — quarterly earnings, management guidance, macro changes, or revised assumptions. Adjusts estimates, recalculates valuation, and flags material changes. Use after earnings, guidance updates, or when assumptions need refreshing. Triggers on "update model", "plug earnings", "refresh estimates", "update numbers for [company]", "new guidance", or "revise estimates".
Build accretion/dilution analyses for M&A transactions. Model pro forma EPS impact, synergy sensitivities, and purchase price allocation. Use when evaluating a potential acquisition, preparing merger consequences analysis for a pitch, or advising on deal terms. Triggers on “merger model,” “accretion dilution,” “M&A model,” “pro forma EPS,” “merger consequences,” or “deal impact analysis”
Catalyst NoSQL — key-value table store with typed items built via NoSQLItem builder. Partition key required. Trigger on 'NoSQL', 'nosql.table', 'insertItems', 'fetchItem', 'updateItems', 'deleteItems', 'queryTable', 'NoSQLItem', 'document storage', 'flexible schema', or 'Catalyst document database'. Do NOT use when you need SQL joins, ZCQL queries, or a fixed relational schema — use catalyst-datastore instead.
Generates MCP server configuration JSON, resolves authentication setup, and validates server connectivity for dbt. Use when setting up, configuring, or troubleshooting the dbt MCP server for AI tools like Claude Desktop, Claude Code, Cursor, or VS Code.
Use when the user asks to “figure out where our direct traffic really comes from”, “measure dark social”, “add a how-did-you-hear-about-us field”, or “show social drives signups without click data”; produces a share-link/UTM hygiene spec for owned share surfaces, a self-reported attribution field design that replaces an existing form field (free-text first, coded later), a GA4 direct-traffic decomposition read (deep-URL directs, mobile-app skew, private-push correlation) with every derived number hard-labeled Estimated/proxy, and a branded-search-lift proxy from GSC plus Wikipedia pageviews — the declared dark-social method behind ECHO O2. Not for paid-channel attribution reconciliation (platform-claimed vs analytics conversions) — use attribution-reconciler. 暗社交归因/直接流量分解/自报来源字段/分享链路UTM
Use when the user asks to “calculate influencer ROI”, “prove campaign value”, or “what was our ROAS”; produces direct ROI/ROAS, earned media value, attribution-modeled revenue, LTV-based ROI, and a stakeholder-ready summary. Not for building the full slide/written report — use report-generator. 达人营销ROI计算/投资回报测算
Catalyst Zia Services — OCR, Face Analytics, Text Analytics, Object Detection, Barcode Reader, and Content Moderation via pre-trained AI APIs. Trigger on 'Zia', 'OCR', 'face detection', 'text analytics', 'object detection', 'barcode', 'content moderation', or 'image recognition'. DC restrictions: Identity Scanner Document Processing is IN DC only (API included); Facial Comparison works via API from any DC (console testing is IN DC only). For QuickML (AutoML, ML pipelines, LLM Serving, RAG) load catalyst-quickml instead.
Use when the user asks to “build email segments from my list”, “make engaged / lapsed / RFM segments”, “set up cart-abandoner or lifecycle-stage audiences”, or “build a suppression list of unsubscribes and bounces”; turns the user's OWN list/CRM/GA4/ecommerce export into behavioral, attribute, and lifecycle-stage segments plus a suppression list, with per-segment sizes labeled Measured/Estimated, informing the SEND E (Engagement/targeting) dimension. Not for scoring EQS or running vetoes — use email-quality-auditor; not for authentication or spam-content checks — use deliverability-qa. 邮件列表分群/生命周期分群/抑制名单/流失召回
Use when the user asks to "set up conversion values so tROAS optimizes profit not orders", "map margin onto my purchase value", "build value rules for lead / phone / signup conversions", or "stop bidding to revenue when I care about profit"; defines and QAs the conversion VALUE model — per-conversion values, margin/net-value adjustment, static-vs-dynamic value rules, proxy values for non-revenue actions, and a value-vs-count sanity check — as a value-model spec plus a pre-launch value QA sheet. Not for whether the tag fires or UTMs are clean — use conversion-signal-qa; not for cross-platform double-count de-dup — use attribution-reconciler; not for scoring R1/R2 — that is a scored veto in ad-account-auditor. 付费广告转化价值建模/利润出价/价值规则QA
[Deprecated] Optional advanced tool for complex data modeling. For simple MySQL table creation, use relational-database-tool directly; for PostgreSQL / CloudBase PG schema work, use postgresql-development. New environments should use PostgreSQL DDL via queryPgDatabase/managePgDatabase—see postgresql-development skill instead
Use when the user asks to "track where my emails are actually landing after I send", "read my seed-list inbox vs spam vs promotions results", "trend my Gmail Postmaster / Microsoft SNDS reputation", or "did placement drop after my last send"; produces a per-provider inbox/spam/promotions placement read, a domain/IP reputation trend from Postmaster + SNDS, a send-over-send delta with named regressions, and a reusable SEND-S placement snapshot on your own exported telemetry. Not for the pre-send SPF/DKIM/DMARC auth pre-flight — use deliverability-qa; not for computing the EQS or running the vetoes — use email-quality-auditor. 邮件收件箱落点监测/收件箱vs垃圾邮件/Postmaster声誉趋势
Use when the user asks to “monitor my launch”, “track our Product Hunt / Hacker News ranking”, or “watch the launch window”; runs the T-0 to T+30 window watch — pre-launch instrumentation verification (UTM/event checks, the upstream of RAMP P1), HN rank/points/comments polling with a comments-over-points flamewar early-warning (Estimated heuristic), PH votes/featured status, store charts and reviews, news echo, D0/W1/M1 KPI snapshots vs targets, spike-vs-sustain and owned-capture reads, and alert thresholds against the launch-tier KPI targets. Not for launch-day go/rollback calls — use launch-day-conductor; not for metric deep-dives — use performance-analyzer; not for SEO rank tracking — use rank-tracker. 发布监控/排名轮询/火焰战比/spike-sustain
Use when creating or modifying dbt Semantic Layer components—semantic models, metrics, dimensions, entities, measures, or time spines. Covers MetricFlow configuration, metric types (simple, derived, cumulative, ratio, conversion), and validation for both latest and legacy YAML specs.
Use when the user asks "what did we pay this creator last time" or to "update the creator roster"; curates creator identity, rate, rights, exclusivity, compliance-event, and performance facts through the append-only creators event stream. Not for scoring fit — use fit-scorer; not for reviewing content — use creator-content-auditor. 创作者档案/达人名册
Use when the user asks to "log this subscriber''s opt-in", record unsubscribes/complaints, or query lawful basis; curates pseudonymous consent facts through the append-only consent stream and applies suppression/erasure tombstones immediately. Not for SEND scoring — use email-quality-auditor; not for building segments — use list-segment-builder. 订阅同意台账/实时退订抑制/合法性依据登记
Use when the user asks to "is my ad fatiguing", "why is CTR dropping at scale", or "should I rotate creative / widen the audience"; reads frequency, CTR and CVR decay against an early-flight baseline and returns Rotate-creative / Widen-audience / Hold triggers with a per-ad-set fatigue read. Not for building the replacement creative — use ad-creative-builder; not for the RQS score or vetoes — use ad-account-auditor. 广告疲劳检测/频次管理/换素材还是扩人群
Use when the user requests mathematical modeling, modeling competitions, modeling analysis, code-based solutions, result visualization, or the creation of a mathematical modeling paper. Supports a complete three-stage workflow, running any role independently, and using an independent Subagent for quality checks within a stage at key points. Only the quality-check Subagent is enabled by default; users can explicitly opt into additional collaboration.
Use when the user asks to "watch my list health over time", "flag decaying / unengaged subscribers on a schedule", "why is my open rate drifting down / bounces creeping up", or "build me a re-permission and prune worklist"; runs the scheduled SEND list-decay + suppression-drift watch — an engagement-recency cohort read (30/90/180/365-day), hard-bounce and spam-complaint trend vs benchmark, suppression-list growth/leakage check, and a segmented re-permission / sunset / prune worklist tied to SEND S (list hygiene) and E (engagement-decay) sub-items. Not for the one-time pre-send authentication pre-flight — use deliverability-qa; not for the consent/suppression record itself — use consent-registry; not for computing the EQS or enforcing vetoes — use email-quality-auditor. 邮件列表健康度监控/退订漂移/沉睡用户清理
Create diverse synthetic test inputs for LLM pipeline evaluation using dimension-based tuple generation. Use when bootstrapping an eval dataset, when real user data is sparse, or when stress-testing specific failure hypotheses. Do not use when you already have 100+ representative real traces (use stratified sampling instead), or when the task is collecting production logs.
Tracks physical units and propagates measurement uncertainty in scientific calculations using pint and uncertainties. Use for unit conversion and dimensional checking, GUM uncertainty budgets, Type A and Type B evaluation, coverage factors and expanded uncertainty, Monte Carlo propagation, significant-figure and plus-minus reporting, error propagation through curve fits, CODATA constants, auditing Python code for stripped units or broken uncertainty propagation, and order-of-magnitude plausibility checks using dimensionless groups (Reynolds, Peclet, Damkohler, Knudsen, Biot, Womersley), characteristic scales such as diffusion time or Debye length, and observed magnitude ranges. Trigger on "is this number physically reasonable", "sanity check these units", "what regime is this flow in", or a result that looks off by orders of magnitude.
Use when the user asks to “run the weekly social readout,” “which denominator does our engagement rate use,” or “which posts won this week and what changes next cycle”; produces the organic-social metric dictionary (every rate names its denominator — ERR engagement-by-reach vs ERI by-impressions vs ER-by-follower — locked across periods), median-not-mean per-post rollups with organic and boosted separated, EMV as labeled exec-translation only (never inside any score), an attributed CHAOSS/Orbit-style community-health readout with employees excluded, and the best/worst-performer write-back the next calendar cycle consumes. Not for dollar-ROI math or the ECHO profile result gate verdict — use roi-calculator and social-quality-auditor. Social weekly report/engagement-rate denominator/metric dictionary/retrospective write-back
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
Use when the user asks to “track our share of voice,” “what share of the conversation do we own vs competitors,” or “trend our SOV this quarter”; computes SOV% = brand mentions ÷ (brand + competitor panel mentions) per platform per period on a LOCKED competitor panel — a panel switch invalidates the trend (restart the series and log the break; the ECHO O3 rule) — plus a sentiment-weighted SOV variant (sentiment labeled Estimated unless human-coded) and a Wikipedia-pageviews attention-share alternative; built from keyless listening connectors, gdelt.py news echo, and user exports — public counts only, closed platforms are never scraped. Not for backlink or offsite SEO signals — use offsite-signal-analyzer. 声量份额/竞品声量对比/提及份额/注意力份额
Use when the user asks to “run a launch retro / post-mortem,” “compare launch results vs targets by channel,” or “decide what to keep or kill for the next launch”; produces a structured D1/W1/M1 retrospective — a per-channel actual-vs-target table (UTM-attributed own analytics as the truth column, platform self-reported numbers as reference, every figure labeled Measured / User-provided / Estimated), a 5-Whys chain on the single largest miss, keep / kill / change decisions per channel, 3-5 actionable learnings for the next launch, and an outcome snapshot submitted to the launch registry. Not for return math (CPA / ROI) — use roi-calculator; not for the stakeholder-facing report writeup — use report-generator; not for a metric deep-dive — use performance-analyzer. Launch retrospective/channel attribution/5 Whys/keep-kill
Use when the user asks to “read back” a paid campaign change, “did this ad change work,” or “compare ROAS/CPA before and after”; reads ROAS/CPA against a control over a fixed readback window and returns a Promote / Keep-testing / Rollback / Unproven readback decision with the math delegated to roi-calculator. Not for RQS scoring or veto adjudication — use ad-account-auditor; not for the ROI ratio math — use roi-calculator; not for cross-channel rollups — use performance-analyzer. Paid campaign retrospective/ROAS readback/ad performance attribution
Use when the user asks to “design an email A/B test,” “set up a multivariate subject/CTA test,” “run a send-time test,” “build a hold-out group,” or “is this email result statistically and practically material?”; produces a falsifiable hypothesis, one-variable-per-cell matrix, sample-size/MDE/duration/power plan, and an effect/uncertainty read from own ESP data. Applies only a precommitted owner-approved action rule; the helper never chooses a business action. Not for EQS/vetoes or writing the email. Email A/B test design/multivariate test/send-time test/hold-out group/significance assessment
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.