# Skills & Agents — Agent Directory You are an agent looking to extend your capabilities. Below is the complete directory of skills and agents available at https://skillsandagents.co. Each entry includes trigger phrases that activate it and any MCP dependencies it needs. - Skills available: 38 - Agents available: 4 --- ## Skills A Skill is a discrete capability you add to Claude. Each one teaches Claude how to do a specific task. ### Ads Copilot - Slug: `ads-copilot` - Category: marketing - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/ads-copilot/ - Tags: marketing, paid acquisition, ads, analytics, optimization, csv Chat with your ad-platform and analytics data, then get 3 to 5 ranked, specific optimization moves with the assumption behind each one A lightweight chat-with-your-ads wedge — not an attribution platform. Ads Copilot discovers whatever sources are reachable in the session (connected MCP servers for Google Ads, Meta, TikTok, LinkedIn, GA4, or Stripe, or CSV exports the user uploads), answers plain-language questions about ad performance first, then surfaces ranked moves with expected effect and the assumption each one rests on. Today, the Anthropic MCP registry has zero ad-platform connectors for the major paid channels, so CSV export is the default data path. The skill treats platform-reported conversions as self-reported and cross-checks against GA4 and Stripe when those are connected. **Trigger phrases:** - "ads copilot" - "how are my ads doing" - "review my ad spend" - "what should I do about my Meta ads" - "optimize my ads" - "audit my ads" - "why isn't growth working" **Required MCP dependencies:** - **Google Ads MCP** — Not currently in the Anthropic MCP registry. CSV export from Google Ads (Reports → Predefined → Campaign performance) is the default path; the skill normalizes the export via scripts/parse_csv.py. ([docs](https://www.anthropic.com/connectors)) - **Meta Ads MCP** — Not currently in the Anthropic MCP registry. CSV export from Meta Ads Manager (Performance and Clicks preset, by Day) is the default path. ([docs](https://www.anthropic.com/connectors)) - **TikTok Ads MCP** — Not currently in the Anthropic MCP registry. CSV export from TikTok Ads Manager (Custom report grouped by Campaign, by Day) is the default path. ([docs](https://www.anthropic.com/connectors)) - **LinkedIn Ads MCP** — Not currently in the Anthropic MCP registry. CSV export from LinkedIn Campaign Manager (Performance report grouped by Campaign) is the default path. ([docs](https://www.anthropic.com/connectors)) - **GA4 MCP** — Google Analytics 4 — used as an independent cross-check on platform-reported conversions. May be available in the registry depending on your setup; CSV export from a Traffic acquisition report works as a fallback. ([docs](https://www.anthropic.com/connectors)) - **Stripe MCP** — Used to validate paying-customer counts and compute CAC payback. May be available in the registry depending on your setup; otherwise the skill skips CAC payback rather than guessing. ([docs](https://www.anthropic.com/connectors)) --- ### Agency Automation Blueprint - Slug: `agency-automation-blueprint` - Category: productivity - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/agency-automation-blueprint/ - Tags: automation, agencies, workflows, operations, claude Runs a guided discovery for agencies and service businesses, maps where work leaks across Gmail, Slack, Monday.com, meetings, and docs, then ranks the automations worth building and hands back a blueprint. Tell this skill how your agency actually works and it runs a guided discovery to find where information leaks between Gmail, Slack, Monday.com, meetings, and documents. It ranks which automations are worth building first, recommends a clean architecture with Claude as the reasoning layer, and hands back a blueprint plus a ready-to-use starter prompt and SOP. Built for service businesses that want practical, reliable workflows, not a pile of brittle automations. **Trigger phrases:** - "help me automate my agency" - "connect my tools with claude" - "my information is scattered" - "reduce manual admin work" - "where should I start with AI automation" --- ### Agent Browser - Slug: `agent-browser` - Category: coding - Version: 5185339 - Author: Vercel Labs - Install type: simple - URL: https://skillsandagents.co/skills/agent-browser/ - Tags: browser, automation, testing, cli, scraping, web Browser automation for AI agents through a fast native Rust CLI: navigate pages, fill forms, click, take screenshots, extract data, and test web apps via accessibility-tree snapshots A third-party skill from Vercel Labs that drives a native Rust browser-automation CLI (agent-browser). Claude calls it through Bash to interact with real websites and Electron desktop apps over the Chrome DevTools Protocol: navigating pages, filling forms, clicking buttons, taking screenshots and PDFs, intercepting network traffic, managing cookies and auth state, and handling multiple tabs. Instead of raw DOM scraping it uses compact accessibility-tree snapshots with stable element references, which makes interactions more reliable. The skill includes a core command set plus specialized add-ons for Electron apps (VS Code, Slack, Discord, Figma), Slack workspaces, QA/testing, and cloud browsers (Browserless, Browserbase, Vercel Sandbox, AWS Bedrock AgentCore). Use it when the user needs to automate a browser, test a web app, scrape or extract data, or reproduce a UI flow. Requires Node.js 24 or newer and installing the pinned agent-browser CLI (npm i -g agent-browser@0.27.2, then agent-browser install, adding --with-deps on Linux), plus Bash access to it. Note the pinned SKILL.md is a discovery stub: the detailed workflows are served by the CLI at runtime via agent-browser skills get core and track the installed version, which is why the npm package is pinned to the reviewed release. Licensed Apache-2.0 by Vercel. **Trigger phrases:** - "automate this browser task" - "fill out this form on the website" - "take a screenshot of this page" - "test this web app" - "scrape data from this site" - "click through this flow" - "/agent-browser" --- ### Buy-Side Due Diligence - Slug: `buy-side-diligence` - Category: finance - Version: 1.1.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/buy-side-diligence/ - Tags: due diligence, acquisitions, quality of earnings, private equity, search funds, financial analysis Runs systematic buy-side due diligence on a private company acquisition: normalizes earnings, stress-tests the seller's narrative, cross-references documents, and quantifies red flags into valuation adjustments. A complete buy-side diligence operating system for private company acquisitions. It deconstructs reported EBITDA into a defensible normalized bridge, stress-tests the balance sheet for hidden liabilities, validates customer concentration and change-of-control exposure, cross-references financials against tax returns and the CIM to surface inconsistencies, and rolls everything into a prioritized red-flag matrix with quantified dollar impact and an investment-committee-ready memo. Built for corporate development teams, PE associates, search fund operators, and M&A advisors. Includes a 30-prompt staged library covering pre-LOI screening through IC outputs, plus a missing-document log that treats the documents the seller did not provide as findings in their own right. It is an analytical accelerator, not a replacement for a sell-side quality-of-earnings report or an accountant's sign-off. **Trigger phrases:** - "run diligence" - "due diligence" - "QoE this" - "normalize this EBITDA" - "build a red flag matrix" - "what's wrong with this CIM" --- ### Cancel Subscriptions - Slug: `cancel-subscriptions` - Category: finance - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/cancel-subscriptions/ - Tags: subscriptions, recurring charges, money, gmail, automation, budgeting Scans your Gmail for recurring charges, maps your active subscriptions, and cancels the ones you don't want Most people are paying for subscriptions they forgot about. This skill scans your Gmail for subscription receipts and billing emails, builds a complete picture of your monthly spend, and then — for any service you want to cancel — navigates to the cancellation page and walks you through the process. Stop paying for things you don't use. **Trigger phrases:** - "cancel my subscriptions" - "what am I paying for every month" - "find my recurring charges" - "help me cut my subscriptions" - "what subscriptions do I have" - "I want to cancel some subscriptions" - "audit my subscriptions" **Required MCP dependencies:** - **Gmail MCP** — Read access to Gmail to scan subscription receipts and billing confirmations ([docs](https://github.com/modelcontextprotocol/servers/tree/main/src/gmail)) - **Computer Use MCP** — Browser automation for navigating cancellation flows on subscription websites ([docs](https://github.com/anthropic-ai/mcp-computer-use)) --- ### Claude Trading Skills - Slug: `claude-trading-skills` - Category: finance - Version: 1.0.0 - Author: tradermonty - Install type: simple - URL: https://skillsandagents.co/skills/claude-trading-skills/ - Tags: trading, investing, stocks, dividends, swing-trading, market-analysis, risk-management, portfolio A Claude Skills toolkit for time-constrained individual investors: structured market review, risk management, trade planning, journaling, and continuous improvement across a core-plus-satellite workflow. Claude Trading Skills is a third-party, open-source toolkit (MIT, by TraderMonty) of around 40 trading skills organized into Core plus Satellite workflows. It is built for long-term investors who hold ETFs and dividend stocks as their core while running disciplined swing trades as a satellite when conditions allow. The goal is not to outsource buy and sell decisions to AI. It structures the work around them: market-regime checks (market-breadth-analyzer, uptrend-analyzer, exposure-coach), portfolio review (portfolio-manager, dividend monitors, trader-memory-core), swing candidate screening (vcp-screener, position-sizer), and trade journaling and review (signal-postmortem, backtest-expert). Recommended starting workflows include a 15-minute daily market check, a weekly portfolio review, and a trade-memory loop. A no-API-key path lets you review conditions, size trades, and journal without paid data subscriptions, though some skills still need public CSVs, chart screenshots, or local files. Many advanced skills can optionally use FMP, FINVIZ, or Alpaca. This is not financial advice, a signal service, or a promise of profitability. **Trigger phrases:** - "/claude-trading-skills" - "run my daily market check" - "review my portfolio" - "screen for swing trade candidates" - "size this trade for me" - "journal this closed trade" - "what's the current market regime" --- ### Code Reviewer - Slug: `code-reviewer` - Category: coding - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/code-reviewer/ - Tags: code review, debugging, security, best practices Reviews code for bugs, security issues, performance problems, and style — with specific, actionable feedback Paste your code and this skill performs a thorough review: logic errors, security vulnerabilities, performance bottlenecks, code smell, and deviations from best practices. Feedback is specific and actionable — not 'this could be better' but 'here's why and here's how to fix it.' **Trigger phrases:** - "review this code" - "what's wrong with this code" - "code review" - "check my code for issues" - "find bugs in this" --- ### Competitor Intel - Slug: `competitor-intel` - Category: research - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/competitor-intel/ - Tags: competitive intelligence, research, strategy, market analysis Researches competitors in real-time — positioning, pricing, recent moves, strengths, and gaps Name a competitor (or your own company's space) and this skill uses web search to pull current information on positioning, pricing, product changes, recent news, and strategic moves. Produces a structured competitive brief you can use for strategy sessions, sales enablement, or investor decks. Requires web search MCP for live data. **Trigger phrases:** - "research my competitors" - "give me intel on" - "what is [company] doing" - "competitive analysis on" - "how does [company] compare to us" **Required MCP dependencies:** - **Brave Search MCP** — Web search access for pulling current competitor information and news ([docs](https://github.com/modelcontextprotocol/servers/tree/main/src/brave-search)) --- ### Content Calendar - Slug: `content-calendar` - Category: content - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/content-calendar/ - Tags: content, planning, social media, editorial, calendar Plans a content calendar for any channel — posts, topics, formats, and publishing cadence for weeks or months ahead Describe your brand, audience, and goals, and this skill generates a structured content calendar complete with topics, formats, angles, and a publishing schedule. Works for social media, newsletters, blogs, YouTube, podcasts, or any content channel. Outputs a calendar you can paste directly into Notion, Airtable, or a spreadsheet. **Trigger phrases:** - "create a content calendar" - "plan my content for" - "help me plan social media posts" - "I need a content strategy" - "what should I post this month" --- ### Create a PRD - Slug: `pm-create-prd` - Category: productivity - Version: f9eaa51 - Author: phuryn - Install type: simple - URL: https://skillsandagents.co/skills/pm-create-prd/ - Tags: product management, prd, product requirements, documentation, product spec Drafts a Product Requirements Document using an 8-section template — summary, contacts, background, objective, market, value props, solution, and release plan An experienced-PM agent that turns a feature idea, research dump, or product brief into a full Product Requirements Document. Walks through the 8 sections that align engineering, design, and leadership — what problem you're solving, who you're solving it for, how you'll measure success, what's in the first release vs later. Asks for any missing context, then produces a structured markdown PRD you can paste into Notion, Linear, or your doc tool of choice. Third-party skill from the productcompass.pm PM Skills Marketplace. **Trigger phrases:** - "write a PRD" - "create a product requirements document" - "draft a PRD" - "I need to write a spec for" - "help me document this feature" - "PRD for" --- ### Customer Interview Script - Slug: `pm-interview-script` - Category: research - Version: f9eaa51 - Author: phuryn - Install type: simple - URL: https://skillsandagents.co/skills/pm-interview-script/ - Tags: product management, customer discovery, user interviews, jtbd, mom test, research Builds a Mom Test-style customer interview script with JTBD probing questions — warm-up, core exploration, wrap-up, plus a note-taking template Drafts a structured customer discovery interview script that surfaces real behavior, not opinions. Follows The Mom Test rules — ask about their past, not your idea; talk less, listen more; never pitch. Generates an opening, warm-up, a 15-20 minute JTBD core exploration (current behavior, pain points, desired outcomes, willingness to pay), probing techniques, and a wrap-up. Adapts to the specific product area, persona, and research objective you provide. Includes a note-taking template for capturing key jobs, biggest pain, current solution, willingness to pay, and surprise findings. Third-party skill from the productcompass.pm PM Skills Marketplace. **Trigger phrases:** - "write an interview script for" - "customer interview questions" - "user research script" - "JTBD interview" - "mom test script" - "discovery interview" --- ### Dispatching Parallel Agents - Slug: `dispatching-parallel-agents` - Category: coding - Version: 6fd4507 - Author: Jesse Vincent - Install type: simple - URL: https://skillsandagents.co/skills/dispatching-parallel-agents/ - Tags: agents, parallelism, orchestration, workflow, productivity Splits work into independent tasks and farms them out to parallel agents when there's no shared state or sequential dependency between them A coordination pattern for when you're facing two or more chunks of work that don't depend on each other. It helps you identify which tasks are genuinely independent, scope each one so an agent can run it without stepping on the others, and fan them out concurrently instead of grinding through them serially. Best paired with Claude Code's subagent tooling. Third-party skill from Jesse Vincent's Superpowers collection. **Trigger phrases:** - "these tasks are independent" - "can we do these in parallel" - "dispatch agents for this" - "fan this work out" - "run these at the same time" --- ### Document Creator - Slug: `document-creator` - Category: productivity - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/document-creator/ - Tags: documents, writing, reports, proposals Generates polished, structured documents from a brief — reports, proposals, memos, and more Tell Claude what you need and a few key details, and it produces a complete, professionally structured document ready to edit or send. Works for reports, proposals, executive memos, project briefs, status updates, and any other document type. No MCPs required — just the skill file and Claude. **Trigger phrases:** - "create a document for me" - "write a report on" - "draft a proposal for" - "I need a memo about" - "help me write a business document" --- ### Email Drafter - Slug: `email-drafter` - Category: comms - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/email-drafter/ - Tags: email, writing, communication, drafting Drafts professional emails from a quick brief — the right tone, length, and structure every time Describe what you need to communicate and who you're writing to, and this skill produces a ready-to-send email. Handles everything from brief follow-ups to detailed client proposals. Calibrates tone automatically based on context — formal for clients, direct for colleagues, warm for intros. **Trigger phrases:** - "draft an email for me" - "write an email to" - "help me write this email" - "I need to email someone about" - "compose an email" --- ### Emil Design Engineering - Slug: `emil-design-eng` - Category: coding - Version: ecf66bb - Author: Emil Kowalski - Install type: simple - URL: https://skillsandagents.co/skills/emil-design-eng/ - Tags: frontend, design, animation, css, ui, motion Encodes Emil Kowalski's design engineering philosophy: UI polish, animation decisions, easing curves, springs, gestures, and the invisible details that make interfaces feel right A third-party skill that turns Claude into a design engineer with Emil Kowalski's craft sensibility. It carries an opinionated animation decision framework (should this animate at all, what is the purpose, what easing, how fast), strong custom easing curves over the weak CSS defaults, spring physics for interruptible gestures, and component rules like scale-on-press feedback, origin-aware popovers, never animating from scale(0), and clip-path reveals. It reviews UI code in a required Before/After/Why table and applies performance rules (only animate transform and opacity, CSS transitions over keyframes for dynamic UI, WAAPI for hardware-accelerated programmatic animation) plus accessibility guards for prefers-reduced-motion and touch hover. Use it when building or reviewing frontend components, animations, drawers, toasts, tooltips, or any interaction where polish matters. Sourced from Emil Kowalski's articles and his animations.dev course. **Trigger phrases:** - "review this animation" - "make this UI feel better" - "what easing should I use" - "design a drawer / toast / tooltip" - "why does this animation feel off" - "polish this component" - "/emil-design-eng" --- ### Expense Tracker - Slug: `expense-tracker` - Category: finance - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/expense-tracker/ - Tags: expenses, budgeting, spending, finance, transactions Pulls transactions from connected accounts, categorizes spending, and produces a clear monthly expense breakdown Connect your bank or credit card via MCP and this skill automatically retrieves your transactions, categorizes them intelligently, flags anomalies, and produces a clean monthly spending report. Understand exactly where your money goes — by category, merchant, and trend — without manual data entry. **Trigger phrases:** - "track my expenses" - "show me my spending" - "where did my money go this month" - "give me an expense report" - "how much did I spend on" **Required MCP dependencies:** - **Plaid MCP** — Connects to bank accounts and credit cards to retrieve transaction data ([docs](https://plaid.com/docs/api/)) --- ### Financial Pulse - Slug: `financial-pulse` - Category: finance - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/financial-pulse/ - Tags: finance, spending, transactions, trends, subscriptions, analysis Analyzes transaction data from any source to produce a categorized 30-day spending breakdown and 3 specific things to look into based on 60-day trends The bank-agnostic analysis engine for your money. Feed it transactions from a bank MCP, a CSV export, or pasted data, and it produces a categorized 30-day spending breakdown, detects your recurring subscriptions, and surfaces 3 specific, actionable things to look into based on 60-day trends — ranked by annualized dollar impact. This skill is the brain. To connect directly to a bank, pair it with one of the Financial Pulse connector agents. **Trigger phrases:** - "financial pulse" - "check my spending" - "where's my money going" - "run the pulse" - "spending review" - "what should I look into" - "money audit" --- ### Finishing a Development Branch - Slug: `finishing-a-development-branch` - Category: coding - Version: 6fd4507 - Author: Jesse Vincent - Install type: simple - URL: https://skillsandagents.co/skills/finishing-a-development-branch/ - Tags: git, branches, merge, pull request, workflow Once the work is done and tests pass, walks through the structured options for integrating a branch — merge, PR, or cleanup — instead of guessing A decision aid for the often-fumbled last mile of a feature branch. When implementation is complete and tests are green, it presents the integration paths clearly — merge directly, open a PR, or clean up and discard — and helps you pick the right one for the situation rather than defaulting. Prevents half-finished merges and orphaned branches. Third-party skill from Jesse Vincent's Superpowers collection. **Trigger phrases:** - "I'm done with this branch" - "how should I merge this" - "wrap up this feature branch" - "ready to integrate this work" - "finish this branch" --- ### Frontend Design - Slug: `frontend-design` - Category: coding - Version: 0075614 - Author: Anthropic - Install type: simple - URL: https://skillsandagents.co/skills/frontend-design/ - Tags: frontend, design, ui, css, react, web design Builds distinctive, production-grade frontend interfaces that avoid generic AI-slop aesthetics — bold typography, cohesive color, motion, and unexpected layouts executed with intent Anthropic's official frontend-design skill. Before writing any code, it commits to a clear aesthetic direction — minimal, maximalist, editorial, brutalist, retro-futuristic, or something true to the context — then implements real, working HTML/CSS/JS, React, or Vue with meticulous attention to typography, color, motion, spatial composition, and atmospheric background detail. It explicitly steers away from the overused fonts (Inter, Roboto, system stacks), cliched purple-gradient-on-white palettes, and cookie-cutter layouts that make AI-generated UIs look generic. Use it whenever you're building or styling a web component, page, dashboard, landing page, or artifact and want the result to feel genuinely designed rather than templated. Third-party skill from Anthropic's official skills repository. **Trigger phrases:** - "build a landing page" - "design a frontend for" - "make this UI look better" - "style this component" - "create a React component" - "beautify this web page" - "build a dashboard UI" --- ### GitHub PR Summarizer - Slug: `github-pr-summarizer` - Category: coding - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/github-pr-summarizer/ - Tags: github, pull requests, code review, automation Reads a GitHub pull request and generates a clear summary of what changed, why, and what reviewers should focus on Point this skill at any GitHub pull request and it produces a plain-English summary of the changes: what was modified, the technical approach taken, potential risks, and specific areas for reviewers to focus on. Useful for async review handoffs, release notes, and onboarding teammates to a complex PR. **Trigger phrases:** - "summarize this PR" - "summarize pull request" - "what does this PR do" - "give me a PR summary" - "explain these changes" **Required MCP dependencies:** - **GitHub MCP** — Read access to GitHub repositories, pull requests, and diffs ([docs](https://github.com/modelcontextprotocol/servers/tree/main/src/github)) --- ### Human Tone - Slug: `human-tone` - Category: comms - Version: 1.1.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/human-tone/ - Tags: writing, tone, editing, humanize Makes Claude write like a real person: plain, direct, no fluff. No banned AI phrases, no marketing language, no performative enthusiasm. Paste any text or give any writing task and this skill rewrites it to sound like a person wrote it, not a marketing department or an AI. Kills banned phrases (seamlessly, leverage, dive into, game-changing), strips em dashes, cuts filler words, and enforces plain direct language. Use it any time the goal is writing that sounds genuinely human. **Trigger phrases:** - "write like a human" - "sound natural" - "don't sound like AI" - "make it sound real" - "write casually" - "no corporate speak" - "humanize this" - "make this less robotic" - "rewrite this naturally" - "drop the fluff" --- ### iMessage History - Slug: `imessage-history` - Category: productivity - Version: 0.1.0 - Author: Darlin Alberto - Install type: simple - URL: https://skillsandagents.co/skills/imessage-history/ - Tags: imessage, messages, macos, search, personal data, archive Searches and summarizes your local macOS iMessage history — read-only, on-device. Mac only (not iOS, Windows, or Android). A community skill that turns your Mac's local iMessage database into something Claude can actually search. Reads ~/Library/Messages/chat.db read-only — nothing leaves your machine, no network calls, no install of dependencies beyond Python 3. Ask things like "when did Alex mention their new address?" or "summarize my texts with Jane last summer" and the skill picks the right command. Capabilities include keyword search across all chats, per-sender stats and activity patterns, context windows around a timestamp, attachment discovery by date or MIME type, and automatic contact resolution against your Address Book. macOS only — does not work on iOS, Windows, or Android. **Trigger phrases:** - "summarize my texts with [contact]" - "when did [person] mention X" - "search my iMessages for" - "show me messages from last summer about" - "find attachments from [contact]" - "/imessage" --- ### Learn Quiz - Slug: `learn-quiz` - Category: coding - Version: a39e7b1 - Author: Thariq Shihipar - Install type: simple - URL: https://skillsandagents.co/skills/learn-quiz/ - Tags: learning, understanding, code review, onboarding, comprehension, teaching Turns Claude into a teacher that quizzes you until you actually understand the work it did, instead of rubber-stamping a diff you never read A teaching overlay for any coding session. Instead of ending with 'done,' Claude becomes a teacher whose only goal is to confirm you deeply understand what changed. It keeps a running checklist covering the problem and why it existed, the solution and its design decisions and edge cases, and the broader impact, and verifies your understanding at each step rather than dumping it all at the end. It has you restate what you think is happening first, fills the gaps, offers ELI5, ELI14, or explain-like-an-intern explanations, and quizzes you with open-ended and multiple-choice questions (correct answer shuffled, not revealed until you submit), showing you code or walking you through the debugger when it helps. The session is not finished until you can demonstrate you understood it. Works best in a harness with an interactive question tool such as Claude Code. Third-party skill, adapted from Thariq Shihipar's public gist; the variant that inspired it was Suzanne's. **Trigger phrases:** - "walk me through this" - "teach me this code" - "make sure I understand" - "explain the changes" - "quiz me on this" - "/learn-quiz" --- ### LLM Visibility Agent - Slug: `llm-visibility-agent` - Category: research - Version: 1.1.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/llm-visibility-agent/ - Tags: ai visibility, llm seo, brand monitoring, editorial planning Reads your weekly AI Visibility report from Amplitude or PostHog, finds the prompts where competitors win and your brand is absent, and hands back 3 to 5 specific posts to close the gaps On-demand editorial intelligence layer on top of the Amplitude or PostHog AI Visibility MCP. Pulls the latest weekly visibility report for your tracked org brand, compares it to prior weeks for trend, scores every gap where a tracked competitor shows up in an LLM answer but your brand does not, and produces a ranked list of recommended posts. Surfaces model-specific gaps (strong in Claude, absent in Google AI Overview), source and citation gaps (which domains the LLMs cite instead), and writes a summary brief to a path you confirm. Pure read and synthesize. Does not crawl LLMs, scrape, or call any model directly. Built for uristocrat.com and generalized so any brand tracked in Amplitude or PostHog AI Visibility can use it. **Trigger phrases:** - "check AI visibility" - "how is my brand showing up in LLMs" - "run the LLM visibility report" - "what AI prompts is my brand missing from" - "AI search visibility brief" - "am I showing up in AI search" - "LLM visibility audit" - "what should I write to rank in ChatGPT" - "why is my brand not in AI answers" - "/llm-visibility" **Required MCP dependencies:** - **Amplitude / PostHog AI Visibility MCP** — Reads weekly AI Visibility reports: scores, topics, prompts, competitors, sources, and per-model rankings for your configured org brand ([docs](https://posthog.com/docs/ai-visibility)) --- ### Local Lead Prospector - Slug: `local-lead-prospector` - Category: marketing - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/local-lead-prospector/ - Tags: sales, prospecting, sdr, lead generation, google places, cold outreach Assembles an outreach list of local businesses from an industry + region pair for SDRs. Returns a CSV with name, category, address, phone, website, rating, and review count, plus a markdown summary. Listing data only, no decision-maker names. Google Places API (New) is the preferred path; a Playwright scrape exists as an opt-in fallback because scraping Google Maps is against Google's Terms of Service. A focused prospecting skill for SDRs. Give it an industry and a region like 'dentists in Austin, TX' or 'HVAC companies in Denver' and it produces a CSV (name, category, address, phone, website, rating, reviews, source_url) plus a markdown summary with contactability counts and a 10-row sample. Two data paths share an identical CSV schema. The preferred path is the official Google Places API (New) with paginated Text Search and a documented field mask, authenticated via the GOOGLE_MAPS_API_KEY environment variable. The fallback path is a Playwright + Chromium scrape of public Google Maps results, which is against Google's Terms of Service and only runs after the user passes an explicit --yes-tos flag in the same turn. The skill pushes back on vague inputs ('Texas' is too broad), refuses pulls above 100 rows with a sub-region-split suggestion, surfaces a cost estimate before any API pull above 50 rows, and translates API errors (billing not enabled, invalid key, quota exceeded) into clear messages instead of tracebacks. Listing data only. No decision-maker enrichment, no personal email lookup, no LinkedIn scraping, no CRM writes. **Trigger phrases:** - "find me dentists in Austin" - "build a lead list for HVAC companies in Denver" - "I need contacts for law firms in Chicago" - "give me a prospecting list" - "who should I cold-call in this market" - "build me an outreach list" - "pull a lead list for [industry] in [city]" --- ### Make Interfaces Feel Better - Slug: `make-interfaces-feel-better` - Category: coding - Version: 3845620 - Author: Jakub Krehel - Install type: simple - URL: https://skillsandagents.co/skills/make-interfaces-feel-better/ - Tags: frontend, design, animation, css, ui, typography Design engineering principles for polished UI: concentric border radius, optical alignment, shadows over borders, interruptible and staggered animations, font smoothing, and tabular numerals A third-party skill based on Jakub Krehel's article 'Details that make interfaces feel better'. It encodes 16 concrete principles Claude applies when building UI or reviewing frontend code: concentric border radius (outer radius equals inner plus padding), optical over geometric alignment, layered transparent shadows instead of borders, interruptible animations via CSS transitions rather than keyframes, split-and-stagger enter animations (~100ms delays), subtle exit animations, font smoothing, tabular numerals to stop layout shift on changing values, and text-wrapping strategies (balance for headings, pretty for body). It also ships a list of 10 common mistakes with fixes and a 14-point review checklist covering nested radius consistency, animation properties, minimum 40x40px hit areas, and avoiding patterns like transition: all. Use it when implementing components, micro-interactions, hover states, shadows, borders, typography, or enter/exit animations. **Trigger phrases:** - "make this interface feel better" - "review this UI" - "fix the border radius" - "improve this component's polish" - "why does this look off" - "audit this frontend code" - "/make-interfaces-feel-better" --- ### Meeting Notes Distiller - Slug: `meeting-notes-distiller` - Category: productivity - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/meeting-notes-distiller/ - Tags: meetings, notes, summaries, action items Turns raw meeting notes or transcripts into clean summaries with decisions, action items, and owners Paste in rough notes, a transcript, or a brain dump from any meeting and this skill extracts what matters: key decisions made, action items with owners, open questions, and a concise summary you can send to attendees. Works on any format — bullet points, stream of consciousness, or full transcripts. **Trigger phrases:** - "distill my meeting notes" - "summarize these meeting notes" - "extract action items from this meeting" - "clean up my meeting notes" - "turn this transcript into a summary" --- ### Meeting Scheduler - Slug: `meeting-scheduler` - Category: productivity - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/skills/meeting-scheduler/ - Tags: calendar, scheduling, meetings, automation Finds the best meeting time across calendars, drafts the invite, and sends it — all in one flow Tell Claude who you need to meet with and how long, and this skill checks calendar availability, proposes times, and creates the calendar invite with all the right details. Connects to Google Calendar or similar via MCP. Eliminates the back-and-forth of scheduling. **Trigger phrases:** - "schedule a meeting with" - "find time to meet with" - "set up a call with" - "book a meeting" - "when can I meet with" **Required MCP dependencies:** - **Google Calendar MCP** — Read and write access to Google Calendar for checking availability and creating events ([docs](https://github.com/modelcontextprotocol/servers/tree/main/src/google-calendar)) --- ### Newsletter Composer - Slug: `newsletter-composer` - Category: content - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/newsletter-composer/ - Tags: newsletter, writing, email, content creation Composes a complete newsletter issue from a brief — opening hook, main content, and sign-off in your voice Describe your newsletter, your readers, and what you want to cover in this issue, and this skill drafts a complete edition: an opening hook that earns the read, the main content with your ideas fleshed out, and a sign-off that leaves readers wanting the next issue. No MCPs required. **Trigger phrases:** - "help me write my newsletter" - "draft a newsletter issue" - "write my newsletter this week" - "compose a newsletter" - "I need to write my newsletter" --- ### Office Tells - Slug: `office-tells` - Category: productivity - Version: 1.0.0 - Author: Brian Aber - Install type: simple - URL: https://skillsandagents.co/skills/office-tells/ - Tags: office, powerpoint, word, excel, document quality, anti-ai Anti-AI quality gate that strips the visual, structural, and metadata tells from every PowerPoint, Word, and Excel file Claude generates A behavioral overlay on top of the pptx, docx, and xlsx skills. It does not replace them, it constrains them. The format skill handles how to build the file; Office Tells strips the signals that mark a document as machine-generated before it ships: pipe separators, em dashes, accent lines under slide titles, beige or default-blue or default-grey color choices, zebra table banding, excessive conditional formatting, Sheet1 names, library-default file metadata, hardcoded totals where a formula belongs, and boilerplate section names. The file type alone is the trigger, no anti-AI wording required. The final step renders the file to images and inspects the metadata, because the visual rules cannot be enforced without actually looking at the output. Running prose voice is delegated to the human-tone skill. **Trigger phrases:** - "make a deck" - "build a pitch deck" - "write a report" - "create a spreadsheet" - "build a financial model" - "make a PowerPoint" --- ### Pre-Mortem Risk Analysis - Slug: `pm-pre-mortem` - Category: productivity - Version: f9eaa51 - Author: phuryn - Install type: simple - URL: https://skillsandagents.co/skills/pm-pre-mortem/ - Tags: product management, risk management, pre-mortem, launch planning, product launch Runs a pre-mortem on a PRD or launch plan — sorts risks into Tigers, Paper Tigers, and Elephants, then classifies each as launch-blocking, fast-follow, or track A pre-mortem agent that imagines your launch failing 14 days from now and works backward to surface what would have gone wrong. Categorizes each risk as a Tiger (real problem, evidence-backed), Paper Tiger (overblown concern), or Elephant (unspoken worry nobody's discussing). Classifies real risks as launch-blocking, fast-follow within 30 days, or track-only. Produces an action plan for every launch-blocking risk with mitigation, owner, and due date. Third-party skill from the productcompass.pm PM Skills Marketplace. **Trigger phrases:** - "run a pre-mortem" - "what could go wrong with this launch" - "pre-mortem on this PRD" - "risk analysis for" - "stress test this launch plan" - "what risks am I missing" --- ### Prioritization Frameworks - Slug: `pm-prioritization-frameworks` - Category: productivity - Version: f9eaa51 - Author: phuryn - Install type: simple - URL: https://skillsandagents.co/skills/pm-prioritization-frameworks/ - Tags: product management, prioritization, rice, ice, opportunity score, kano, frameworks Reference guide to 9 prioritization frameworks — RICE, ICE, Opportunity Score, Kano, MoSCoW, Weighted Decision Matrix, and more — with formulas and when-to-use guidance A reference skill for picking and applying the right prioritization framework for your context. Covers 9 methods: Opportunity Score (Dan Olsen's Importance × (1 − Satisfaction) for prioritizing customer problems), RICE, ICE, Kano Model, MoSCoW, Eisenhower Matrix, Impact vs Effort, Risk vs Reward, and Weighted Decision Matrix. Each entry includes the formula, the situation it fits, and links to working spreadsheet templates. Use it when you're comparing RICE vs ICE, deciding how to score a backlog, or aligning a team on what to build next. Third-party skill from the productcompass.pm PM Skills Marketplace. **Trigger phrases:** - "which prioritization framework should I use" - "RICE vs ICE" - "how do I prioritize this backlog" - "prioritization framework" - "opportunity score" - "kano model" --- ### Receiving Code Review - Slug: `receiving-code-review` - Category: coding - Version: 6fd4507 - Author: Jesse Vincent - Install type: simple - URL: https://skillsandagents.co/skills/receiving-code-review/ - Tags: code review, feedback, quality, collaboration, engineering Processes code-review feedback with technical rigor — verify each point before implementing, push back on what's wrong, no performative agreement A discipline for handling review comments well. Instead of blindly implementing every suggestion or nodding along, it pushes you to actually verify each point: is the feedback correct, does it apply, what's the evidence. Unclear or technically questionable notes get interrogated rather than rubber-stamped. The result is changes you can defend, not just changes that close the thread. Third-party skill from Jesse Vincent's Superpowers collection. **Trigger phrases:** - "I got code review feedback" - "help me respond to this review" - "work through these review comments" - "address this PR feedback" --- ### Research Brief - Slug: `research-brief` - Category: research - Version: 1.0.0 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/research-brief/ - Tags: research, analysis, briefing, knowledge Synthesizes a topic into a structured research brief — background, key players, open questions, and recommended reading Give Claude a topic and it produces a thorough research brief: background context, key facts and figures, major players or perspectives, current debates, open questions, and a curated reading list. Useful for getting up to speed quickly on an unfamiliar subject before a meeting, pitch, or decision. **Trigger phrases:** - "give me a research brief on" - "brief me on" - "I need to understand" - "help me get up to speed on" - "what do I need to know about" --- ### Unemployment Guide - Slug: `unemployment-guide` - Category: finance - Version: 1.0.1 - Author: uristocrat - Install type: simple - URL: https://skillsandagents.co/skills/unemployment-guide/ - Tags: unemployment, benefits, job loss, layoff, government, life admin Walks you through filing for unemployment in your US state — agency info, application steps, required documents, weekly certification rules, and the pitfalls that get claims denied A free skill for anyone who just lost their job. Tell it your state and it gives you the official unemployment agency, the online filing URL, the claimant phone number, the documents you need to gather, the application steps, and the weekly (or biweekly) certification rules — tailored to that state. Covers all 50 states plus DC. Flags state-specific gotchas like Florida's required Employ Florida work registration (claims aren't paid until it's complete), Texas's 3-day WorkInTexas rule, California's EDD ID.me verification, and dependent allowances. Warns about the things that get claims denied: how to describe a layoff vs a firing, severance offsets, 1099 misclassification, missed certifications, and the multi-state rule of contacting the state where you most recently worked rather than where you live. **Trigger phrases:** - "I just got laid off" - "apply for unemployment" - "file for unemployment" - "I lost my job how do I file" - "unemployment claim" - "how do I get unemployment benefits" - "/unemployment" --- ### User Stories - Slug: `pm-user-stories` - Category: productivity - Version: f9eaa51 - Author: phuryn - Install type: simple - URL: https://skillsandagents.co/skills/pm-user-stories/ - Tags: product management, user stories, agile, backlog, acceptance criteria Breaks a feature into INVEST user stories with the 3 C's — Card, Conversation, Confirmation — and clear acceptance criteria for each Turns a feature spec or design link into a backlog of well-formed user stories. Follows the 3 C's framework (Card, Conversation, Confirmation) and INVEST criteria (Independent, Negotiable, Valuable, Estimable, Small, Testable). For each story, generates a title, an 'As a [role], I want to [action], so that [benefit]' description, a design link, and 4-6 testable acceptance criteria including edge cases and accessibility considerations. Third-party skill from the productcompass.pm PM Skills Marketplace. **Trigger phrases:** - "write user stories for" - "break this feature into stories" - "create a backlog for" - "user stories" - "acceptance criteria for" - "split this feature" --- ### Using Git Worktrees - Slug: `using-git-worktrees` - Category: coding - Version: 6fd4507 - Author: Jesse Vincent - Install type: simple - URL: https://skillsandagents.co/skills/using-git-worktrees/ - Tags: git, worktrees, workflow, isolation, engineering Sets up an isolated workspace for feature work via git worktrees (or native equivalents) so new work never collides with your current checkout A workspace-isolation helper for parallel or risky work. Before starting a feature or executing a plan, it ensures you're working in an isolated environment — a git worktree or native equivalent — so experiments and long-running branches don't disturb your main checkout. Makes it cheap to run several streams of work side by side. Third-party skill from Jesse Vincent's Superpowers collection. **Trigger phrases:** - "set up a worktree for this" - "I need an isolated workspace" - "use git worktrees" - "work on this without touching main" --- ### Verification Before Completion - Slug: `verification-before-completion` - Category: coding - Version: 6fd4507 - Author: Jesse Vincent - Install type: simple - URL: https://skillsandagents.co/skills/verification-before-completion/ - Tags: verification, quality, testing, discipline, engineering Bars you from claiming work is done, fixed, or passing until you've run the check and confirmed the output — evidence before assertions, always A guardrail against premature success claims. Before saying something is complete, fixed, or passing — and before committing or opening a PR — it requires you to actually run the verification command and read the result. No 'should work,' no assuming green. The rule is evidence first, claim second. Third-party skill from Jesse Vincent's Superpowers collection. **Trigger phrases:** - "is this actually done" - "verify before I commit" - "confirm this is passing" - "check the work is really fixed" - "prove it works" --- ## Agents An Agent is an autonomous worker that runs end-to-end on a defined job. Agents can use Skills, call external tools, and report back when done. ### Financial Pulse — Grasshopper - Slug: `financial-pulse-grasshopper` - Category: finance - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/agents/financial-pulse-grasshopper/ - Tags: finance, grasshopper, narmi, banking, mcp, spending, business Connects to your Grasshopper Bank account via its MCP server, pulls transactions, and runs the Financial Pulse — categorized spending and 3 things to look into A connector agent for Grasshopper Bank business clients. It connects to the Grasshopper MCP server (built by Narmi — the first MCP server shipped by a U.S. bank), pulls 60 days of accounts, balances, and transactions read-only, then runs the Financial Pulse analysis: a categorized 30-day spending breakdown, subscription detection, and 3 business-aware recommendations ranked by annualized dollar impact — duplicate vendor payments, SaaS sprawl, vendor concentration risk, and more. **Trigger phrases:** - "financial pulse" - "check my Grasshopper spending" - "check my Grasshopper account" - "how's my business spending" - "run the pulse" **Required MCP dependencies:** - **Grasshopper Bank MCP** — Read-only access to Grasshopper Bank accounts, balances, transactions, and expense categorization. Built by Narmi. Connected with a bearer token and header secret generated in the Grasshopper Control Center and added to Claude Desktop's config. ([docs](https://support.grasshopper.bank/hc/en-us/articles/40312349814164-Claude-Integration)) --- ### Financial Pulse — Mercury - Slug: `financial-pulse-mercury` - Category: finance - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/agents/financial-pulse-mercury/ - Tags: finance, mercury, startup, banking, mcp, spending, burn-rate Connects to your Mercury Bank account via its official hosted MCP server, pulls transactions, and runs the Financial Pulse — categorized spending, burn rate, and 3 things to look into A connector agent for Mercury Bank, the business bank for startups and SMBs. It connects to Mercury's official hosted MCP server over OAuth — no API keys to manage — pulls 60 days of accounts, transactions, cards, and treasury data read-only, then runs the Financial Pulse analysis: a categorized 30-day spending breakdown, subscription detection, and 3 startup-aware recommendations including burn rate trajectory, months of runway, SaaS sprawl, and card-level anomalies. **Trigger phrases:** - "financial pulse" - "check my Mercury spending" - "check my Mercury account" - "how's my burn rate" - "run the pulse" **Required MCP dependencies:** - **Mercury MCP** — Mercury's official hosted MCP server. Read-only access to accounts, transactions, cards, recipients, and treasury via OAuth — Mercury runs the server, no local install. ([docs](https://mercury.com)) --- ### Financial Pulse — Ramp - Slug: `financial-pulse-ramp` - Category: finance - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/agents/financial-pulse-ramp/ - Tags: finance, ramp, corporate-card, expenses, mcp, spending, departments Connects to your Ramp corporate card account via its official MCP server, pulls card transactions and expense data, and runs the Financial Pulse — spending by department, cardholder anomalies, and 3 things to look into A connector agent for Ramp, the corporate card and spend management platform. It connects to Ramp's official open-source MCP server, which loads your full spend feed (card transactions, reimbursements, and bills) plus departments, vendors, and users into an in-memory database, then queries it with SQL. The agent runs the Financial Pulse analysis: a categorized 30-day spending breakdown, subscription detection, and 3 corporate-card-aware recommendations including department spend variance, cardholder anomalies, and merchant category overlap. **Trigger phrases:** - "financial pulse" - "check my Ramp spending" - "check my Ramp account" - "how's our corporate spend" - "expense review" - "run the pulse" **Required MCP dependencies:** - **Ramp MCP** — Ramp's official open-source MCP server (ramp-public/ramp_mcp). Loads corporate card and expense data into an ephemeral in-memory SQLite database for SQL querying. Self-hosted; requires Ramp Developer API credentials. ([docs](https://github.com/ramp-public/ramp_mcp)) --- ### Ghost SEO Agent - Slug: `ghost-seo-agent` - Category: content - Version: 1.0.0 - Author: uristocrat - Install type: mcp-powered - URL: https://skillsandagents.co/agents/ghost-seo-agent/ - Tags: seo, ghost, blogging, google search console, metadata, indexing Pulls non-indexed URLs from Google Search Console, diagnoses each one, applies metadata fixes to Ghost posts, and submits re-indexing requests — fully autonomous A free agent for Ghost publishers. Logs into Google Search Console, pulls your non-indexed and excluded URLs, cross-references them against your Ghost posts, and writes metadata fixes directly via the Ghost MCP — meta descriptions, excerpts, meta titles, and members-only teasers. Then submits re-indexing requests for every post it touches. No CSV exports, no clicking through dashboards. **Trigger phrases:** - "run SEO audit for my Ghost site" - "audit my Ghost site SEO" - "fix my Ghost indexing" - "check why my posts aren't indexed" - "run the Ghost SEO skill" - "ghost seo audit" **Required MCP dependencies:** - **Ghost MCP Server** — Connects Claude to your Ghost Admin API so it can read and update posts, metadata, excerpts, and tags ([docs](https://github.com/jgardner04/ghost-mcp-server)) ---