Workflow for optimizing ClickHouse and HogQL queries. Use when a HogQL query, query runner, insight, or report is too slow; when a hand-written ClickHouse query (via `sync_execute` or in a migration) is too slow; when ClickHouse times out or hits memory limits; when investigating a slow `system.query_log` row; or when reviewing a proposed HogQL printer change for performance. Covers extracting the ClickHouse SQL, common smells (`FROM ... FINAL`, `JSONExtract` over properties, missing skip indexes, self-joins, CTE blow-up), measuring against a real cluster, and applying the fix at the right layer (printer, query runner, or migration). Does NOT cover Postgres / Django ORM / app-database queries; for those use `profiling-slow-api-endpoints`.

Create a recurring AI-generated PostHog report — schedule a free-text prompt to run on a cron, with the LLM-synthesized markdown delivered to email, Slack, or Microsoft Teams on each tick. Use when the user wants a recurring AI summary of X on any cadence (daily, weekly, monthly, yearly) rather than a one-off report. (To attach an AI summary to an existing insight/dashboard subscription instead of a free-text prompt, see `managing-subscriptions` and its `summary_enabled` option.)

Set up Watt in Claude, or fix it when it stops working. A guided check that confirms you're on a supported surface, verifies the Signal Graph connection and file downloads for you, and walks you through turning on automatic updates — fixing anything that's off, one step at a time. Run it the first time you install Watt, or any time something's not connecting — "set up Watt", "where do I start", "connect Watt", "Watt isn't working", "my export is blocked", "the connector's grayed out" — or type /watt:configure.

Review copy as an intended reader encountering it for the first time, with no access to the writer's unstated context. Identify confusing language, missing setup, assumed knowledge, broken expectations, trust-eroding or off-putting moments, and points where the reader may stop. Use for cold reads, fresh-eyes reviews, first-time-reader perspectives, reader-journey checks, or feedback on what feels confusing, missing, alienating, or hard to follow.

When the user wants to reduce churn, build cancellation flows, set up save offers, recover failed payments, or implement retention strategies. Also use when the user mentions 'churn,' 'cancel flow,' 'offboarding,' 'save offer,' 'dunning,' 'failed payment recovery,' 'win-back,' 'retention,' 'exit survey,' 'pause subscription,' or 'involuntary churn.' This skill covers voluntary churn (cancel flows, save offers, exit surveys) and involuntary churn (dunning, payment recovery). For post-cancel win-back email sequences, see email-sequence. For in-app upgrade paywalls, see paywall-upgrade-cro.

Run an application-wide, read-only architecture audit that finds materially useful simplifications in a codebase's data structures, state representation, control flow, algorithms, and ownership boundaries. Fans out bounded read-only agents per subsystem, verifies every finding against the repo, and ranks results P0 to P3. Use when asked to audit architecture, review how state or data is modelled, find structural simplifications, or invoke /audit-architecture. Not a style or line-level quality pass (use /simplify for that) and not a bug hunt (use /code-review).

Extract brand identity (fonts, colors, typography, spacing, favicons) from any URL and install fonts in an AEM Edge Delivery Services project via Adobe Fonts (Typekit) or Google Fonts. Detects fonts using 4 browser API layers (document.fonts, CSS import-rule walk, Performance API, computed style voting), then resolves via a 5-step cascade: system font, Typekit kit match, Adobe Fonts add-to-kit, Google Fonts fallback, or not-found. Updates head.html with font embed tags. Use when starting a site migration, setting up an EDS project, or extracting a site's visual identity. Triggers on: brand setup, brand extract, brand fonts, install fonts, configure brand.

Orchestrate the full agentic-engineering chain on one brief — map → research → story → spec → rubric → backend lane → validator → frontend lane → final validator → PR — with three human checkpoints (story, spec+rubric, PR). Default path is in-conversation: each step runs as a forked Agent subagent so output is visible inline and per-step model selection is possible (Haiku for read-heavy steps, Sonnet/parent for writers, parent for code). Headless escape hatch: bash-invoke ${CLAUDE_SKILL_DIR}/bin/feature.sh for true AFK runs that survive terminal close. Triggers: /feature, "build this feature", "run the full chain", "orchestrate this feature end-to-end".

Type registry, schemas, and content guidance for development artifacts — projects, epics, stories, tasks, specs, ADRs, spikes, work-items, workstreams, milestones. Load when authoring a new artifact, validating one, or resolving the type system. For simple reads (get/list/find), read the artifact file directly without loading this skill.

Workshop coach for the Research Desk (SEC agents) workshop. Use when the user types /workshop, asks for a workshop act or module ("act 2", "next act", "where am I"), wants a TODO(workshop-N) implemented or explained, or asks for help following WORKSHOP.md.

Interactively walk a user through PR review comments one at a time, collect a per-comment decision, then post agent-authored replies on GitHub and resolve the review threads once the user approves a preview. Use only when the user wants to reply to or resolve review threads on GitHub. Skip when the user only wants comments fetched or displayed (use `pr-comments`), or only wants the code changes made without posting anything back to GitHub.

Design, analyze, and document A/B tests for conversion, onboarding, pricing, lifecycle, and product experiments. Use when the user asks for `/ab-test-setup`, experiment design, sample size, statistical significance, A/B test analysis, ICE-scored test backlogs, or avoiding common testing mistakes.

Load when building, running, testing, packaging, linting, or configuring a Kotlin/Java project with the Kotlin Toolchain (JetBrains' unified CLI, formerly Amper), when scaffolding a new or greenfield Kotlin project, or when the repo has project.yaml, module.yaml, or a ./kotlin wrapper. Skip for existing Gradle/Maven projects.

Generate unit tests for a .NET service following the coverage-kit conventions. Use when the user asks to 'generate tests', 'backfill tests', 'characterize this service', or 'add tests for' a target after coverage-init has run. Operates in characterization mode (freeze current behavior for existing code) or spec mode (assert intended behavior for new/changed code), enforces the run-capture-fill loop, routes untestable units to the manifest cannot-test log, can fan the backfill out across a user-chosen number of parallel agents for large worklists (asking first, since more agents cost more tokens), and runs a single read-only suite critique before the baseline locks in (auditing cannot-test legitimacy, assertion/C1 depth, systematic gaps, and any frozen defect missing from the manifest latent_bugs backlog, never the C0/C1 numbers themselves).

Run a Gradle project's tests by test type or module — unit, integration, or all. Use this skill whenever the user wants to run, re-run, or check the project's tests (e.g. "run the unit tests", "run all tests", "run the integration tests for the vet module", "did the tests pass?").

Run a rigorous, research-backed evaluation of a startup or business idea and produce a scored analysis.json, a polished standalone HTML report, and a searchable index of all evaluations. Use whenever the user wants to evaluate, score, vet, compare, or stress-test a startup idea, business opportunity, product concept, side-project bet, or asks "is this worth building / pursuing?" - also when they invoke /evaluate-startup or mention opportunity scouting, idea evaluation, or their startup-ideas cauldron. Runs an adaptive founder interview, parallel deep research, adversarial fact-checking, and renders results with the bundled render.py script. Do NOT use for reviewing existing code, valuing public companies, or general market research that needs no go/no-go verdict.

Design command-line interface parameters and UX: arguments, flags, subcommands, help text, output formats, error messages, exit codes, prompts, config/env precedence, and safe/dry-run behavior. Use when you're designing a CLI spec (before implementation) or refactoring an existing CLI's surface area for consistency, composability, and discoverability.

Parallel read-only multi-agent root-cause investigation for bugs, regressions, crashes, flaky behavior, or unexplained failures. Use when the user asks to investigate a bug, find the root cause, trace a regression, understand why something broke, or wants a ranked diagnosis with the fastest proof path without making code edits.

Premium UI/UX design auditor with Jobs/Ive philosophy. Use when reviewing, evaluating, or elevating existing UI — triggers on "audit my design", "review UI", "make it feel premium", "design review", "UX audit", "elevate the design", "Jobs/Ive style", or when asked to improve visual quality of an existing interface without changing functionality.

Execute read-only SQL queries against multiple Microsoft SQL Server databases. Use when: (1) querying MSSQL/SQL Server databases, (2) exploring database schemas/tables, (3) running SELECT queries for data analysis, (4) checking database contents. Supports multiple database connections with descriptions for intelligent auto-selection. Blocks all write operations (INSERT, UPDATE, DELETE, DROP, etc.) for safety.