Guide for writing PostHog agent skills — job-to-be-done templates that teach agents how to use MCP tools to achieve a goal. Use when adding new product functionality that agents should know how to work with, creating a new skill, or updating existing skills in products/*/skills/ or .agents/skills/.

Read who a built audience reaches when the signals are already in hand — a signal stack fresh from generate, an explore pool, a pasted audience record, or signals the user names — skip discovery, materialize, and render the two-section read (your signals + discovered). The signals-in-hand way into audience-analyze, behind /watt:audience. Aggregates only — never individual records, never an export. Not a user command. Use when a built audience is in session or the user supplies its signals and asks who's in it — "who's actually in this audience", "what do these people look like".

Send emails, buy forwarding inboxes, manage custom subdomains, and read messages programmatically via x402. USE FOR: - Sending emails (one-off or from custom addresses) - Buying a forwarding inbox (username@stableemail.dev) - Purchasing custom email subdomains (you@yourname.stableemail.dev) - Reading and managing inbox messages programmatically - Setting up catch-all forwarding for subdomains - Managing authorized signers for shared subdomains TRIGGERS: - "send email", "email this to", "notify by email" - "buy inbox", "forwarding address", "email inbox" - "custom domain", "subdomain", "email subdomain" - "read emails", "check inbox", "list messages" - "email address", "disposable email", "temporary email" ALWAYS use agentcash.fetch for stableemail.dev endpoints. IMPORTANT: Use exact endpoint paths from the Quick Reference table below.

When the user wants to find co-marketing partners, plan joint campaigns, or brainstorm partnership opportunities. Use when the user says 'co-marketing,' 'partner marketing,' 'joint campaign,' 'who should we partner with,' 'integration marketing,' 'cross-promotion,' 'collaborate with another company,' 'partnership ideas,' or 'co-brand.' For customer referral programs, see referrals. For launch-specific partnerships, see launch.

When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program," or "experiment playbook." Use this whenever someone is comparing two approaches and wants to measure which performs better, or when they want to build a systematic experimentation practice. For tracking implementation, see analytics. For page-level conversion optimization, see cro.

DeFi yield research and liquid staking via EmblemAI. Discover yield opportunities, compare protocols, check DeFi positions with Nansen, and enter liquid staking via token swaps. Use when the user wants to research yields, find staking options, or review DeFi positions.

Track crypto portfolio across 7 blockchains via EmblemAI. Aggregated balances with USD values, conditional trade P&L, and DeFi position tracking via Nansen. Use when the user wants to check their portfolio, see balances across chains, or review trade positions.

Guide for creating coding standards via the Packmind CLI. This skill should be used when users want to create a new coding standard (or add rules to an existing standard) that captures team conventions, best practices, or coding guidelines for distribution to Claude.

SkillsBench task PR review — classifies the task track (standard / research / multimodal), runs static policy checks against the track-specific rubric, benchmarks the task across oracle plus Claude and Codex (with and without skills), audits trajectories for cheating and skill invocation, and produces a `pr-N-task-timestamp-run.txt` review report alongside a `prN.zip` bundle of trajectories. Use when reviewing a SkillsBench task PR (by number, branch, or local task path), when the user asks to review a task, run benchmarks on a PR, audit a submission, classify a task as research or multimodal track, or prepare a comment to post on a SkillsBench PR.

Summarize long Isabelle or Coq proof scripts into high-level logical steps and reasoning flow. Use when users need to: (1) Understand the structure of a complex proof, (2) Document proof strategies for others, (3) Extract the key reasoning steps from verbose proof scripts, (4) Create readable proof outlines from detailed tactical proofs. Produces hierarchical outlines with moderate detail showing proof structure, main cases, key lemmas, and reasoning flow for both Isabelle/Isar and Coq proofs.

Analyze MCP server usage instrumented with Amplitude's MCP Analytics SDK: break usage and errors down by tool, read the rationales within each tool to see what callers are trying to do, and produce a prioritized write-up of actionable fixes. Use this skill whenever the user asks to understand how their MCP server is being used, what agents/users are trying to do with it, why tool calls are failing, what to fix or improve in their MCP server, or asks for an "MCP usage report", "tool error analysis", "intent analysis", "rationale clustering", or "MCP insights". Also trigger when the user mentions [MCP]-prefixed events, tool rationale, tool call errors, or just finished instrumenting their MCP server and wants to see what the data says. Requires the Amplitude MCP connector.

Generate a static interactive D3 walkthrough of a pull request. Use when the user wants a zoomable PR map, graph/canvas PR orientation, or alternate visualization of PR system components, data flow, code dependencies, and user actions.

Use SpriteCook saved presets through MCP. Use when the user explicitly asks to use a preset, refers to one of their saved SpriteCook presets, asks for the same style/settings as a saved preset, or asks to save a private preset for future consistent asset generation.

Produce the latest deterministic C0/C1 coverage report for a .NET service. Use when the user asks for a 'coverage report', 'coverage status', 'what's covered', 'C0 C1 numbers', or wants to check the gate. Runs dotnet-coverage + ReportGenerator (numbers are tool output, never estimated), then joins uncovered files against the manifest to explain every inclusion, exclusion, and cannot-test entry. Runnable any time.

Installs the Amplicode IntelliJ plugin into IntelliJ IDEA (Ultimate/Community) and GigaIDE on the user's machine. Trigger explicitly when the user asks to install the Amplicode plugin into their IDE, in any language. Also trigger implicitly when another Spring Agent Toolkit skill (`spring-explore`, `spring-data-jpa`, `spring-data-jdbc`, `crud-rest-controller`, `dto-creator`, `mapper-creator`, `spring-security-configuration`, `spring-planning`, `java-debug`) detects that the Amplicode MCP server is not connected and delegates installation to this skill. After the IDE-plugin is installed, the user must open any project and click the "Настроить Spring Agent" button on the welcome screen — that is what triggers MCP auto-configuration and spring-skills install. This skill does NOT configure MCP itself. Skip GoLand, PyCharm, WebStorm, Rider, CLion, RubyMine, PhpStorm, DataGrip and other JetBrains products — only IDEA and GigaIDE are supported targets.

Control the user's currently open Chrome tab through the Playwriter CLI (no new browser launch). Use when you need to inspect live UI state, run scripted browser actions, capture console output, or reproduce frontend issues directly in the user's tab.

Guide a user end-to-end through setting up Chrome Web Store API release automation in any repository. Use when asked to walk someone through OAuth/CWS credential setup, refresh token creation, local/CI secret setup, version-based publish automation, and submission status checks.

Build, refactor, or review macOS menubar apps that use Tuist and SwiftUI. Use when creating or maintaining LSUIElement menubar utilities, defining Tuist targets/manifests, implementing model-client-store-view architecture, adding script-based launch flows, or validating reliable local build/run behavior without Xcode-first workflows.

Convert scanned PDFs and document images into clean Markdown using docling for layout (figures, tables, reading order) plus a vision-language OCR model. Use when a user needs high-quality OCR of scanned documents, historical literature, or photographed pages — preserving multi-column reading order, diacritics, special characters, and figures. Supports local vLLM/Ollama servers and cloud vision APIs (OpenAI, Anthropic). Assumes an OCR backend already exists.

Process receipts, track expenses in Google Sheets, reconcile records, and generate entertainment supplement tables for Field Museum procurement card accounting