Claude Skills by gethamster

Design, implement, review, and repair model-facing AI SDK tools in apps/ai, including Zod input schemas, field descriptions, JSON Schema conversion, validation-error loops, tool-call repair hooks, and contract tests. Use whenever adding or changing an agent tool, diagnosing tool input validation failures, or reviewing whether a tool gives the model enough guidance.

Analytics tracking patterns for Segment, Mixpanel, and user engagement metrics. Use when implementing event tracking, user properties, or engagement counters.

Context graph performance validation for Hamster Studio. Use when changing context graph RPCs, indexes, RLS/security modes, live-docs search, code search / search_code_with_neighbors behavior, or SQL migrations that can affect query plans.

Core Hamster Studio conventions for TypeScript, security, and project structure. Always-on essentials that apply across all development tasks. Reference when writing any code in the monorepo.

Investigate main-thread freezes / slow pages with a real CPU profile instead of guessing. Use when a page locks up, the sidebar/cards go unresponsive, requests hang pending, a 'freeze' or 'jank' is reported, or you need to prove a perf fix worked (before/after). Drives a real browser via agent-browser, captures a Chrome DevTools profile, and parses it for the dominant hot frame — separating real app cost from profiler/dev-mode artifacts.

Database testing patterns for Hamster Studio. Covers both Vitest (RLS policies, constraints via Supabase client) and pgTAP (triggers, functions, schema assertions via SQL). Use when writing, reviewing, or debugging any database test — this skill helps you pick the right tool.

Supabase database patterns for Hamster Studio. Use when working with schemas, migrations, RLS policies, SQL functions, or database queries. Essential for any database-related work.

Prove a claim about sandbox-side behavior by running it in a real E2B sandbox instead of reading comments. Use when a delivery or environment boot fails inside the sandbox, when a probe/health-check times out or hangs, when you need to know what is actually installed in a snapshot and at what version, when a comment asserts how a third-party CLI behaves, or when someone proposes raising a timeout. Covers the throw-shapes E2B uses for exit-vs-timeout, cold snapshot-resume timing, deliberate br...

Deterministic realtime E2E testing patterns for Supabase postgres_changes. Use when writing E2E tests that seed data via DB and expect it to appear in the browser via Supabase Realtime.

Add and manage feature flags for gating features in Hamster Studio across environments and integrations.

Plan-time observability for new features in Hamster Studio. Use PROACTIVELY during plan/brief generation, feature scoping, architectural design, or anytime a new endpoint, server action, BullMQ job, page route, or user-facing surface is being designed. Forces every plan to declare PostHog events, user/account properties, group context, feature flags, session-replay/PII posture, and trace correlation BEFORE code is written. Complements `hamster-analytics` (the implementation API) by making ins...

External integration patterns including Airweave context connectors, OAuth flows, and third-party service connections. Use when implementing new integrations or working with context connections.

Keep MCP (apps/mcp) and CLI (apps/cli) in sync with web when a new Studio artifact/entity or agent-facing operation is added. Use when web (apps/web) gains an entity/table/capability agents should reach, or when editing packages/go/ops, apps/mcp, apps/cli, or parity_manifest.json. Triggers: 'new entity/artifact in web', 'expose X to agents/MCP/CLI', 'add an MCP tool', 'add a CLI command', 'parity_manifest', 'parity test failing'.

Rollout safety rules for every change under apps/web/supabase/migrations/**. Use whenever you write, review, rename, or recover a Supabase migration — especially in pre-PR review or when a deploy broke an environment. Migrations install schema and fail closed; production data repair, backfill, reconciliation, and CREATE INDEX CONCURRENTLY belong in explicit audited scripts run before/after deployment. Trigger even if the user only says 'add a column', 'fix these legacy rows in the migration',...

NestJS backend patterns for apps/ai and apps/sync. Use when building background jobs, AI pipelines, sync processing, queues, or working with the shared @kit/api-core package.

PR checklist for NestJS module and dependency-injection changes in apps/ai, apps/sync, and packages/api/core. MUST USE when a PR changes *.module.ts, AppModule, module imports/providers/controllers/exports, @Injectable constructor dependencies, custom provider tokens, dynamic modules, or @Global modules.

Next.js API route handlers in apps/web. Use when building CRUD endpoints, data-serving API routes (app/api/), webhooks, or OAuth callbacks in the Next.js frontend. NOT for background jobs or AI pipelines (use hamster-nest-api instead).

Use this skill whenever the user asks to review, analyse, investigate, duplicate, or build a PostHog dashboard for Hamster Studio. Drives the workflow that turns a dashboard ID (or a request to build one) into a snapshot, an analysis report, and prioritised follow-ups. Captures the doc structure (`posthog-reports/`, `posthog-snapshots/`, `posthog-insights-scope.md`, `posthog-insights-progress.md`, `posthog-events-inventory.md`) and the scripts in `packages/glass-box/scripts/`.

Per-user PostHog event audit for Hamster Studio. Use when the user asks to \\\"review my events in posthog\\\", \\\"audit attribution for <user>\\\", \\\"tell me the story of what user <X> did\\\", \\\"verify $ai_generation events for <user>\\\", or \\\"did event Y fire when I clicked Z\\\". Two modes — guided (ask the user to perform a sequence of actions, then audit that exact window) and free (audit whatever happened in a recent window and reconstruct the story). Sister to `hamster-posthog...

Self-review checklist run before opening a PR. Validates skill compliance, Supabase admin client usage, migration discipline, dead code, DRY, HA suitability, and NestJS DI hygiene. Apply right before `gh pr create`, after staging the final diff.

Hamster-specific React and Next.js patterns. Use when creating pages, components, forms, server actions, or data fetching in apps/web. Complements the react-best-practices skill.

Deterministic integration testing patterns for `apps/sync`. Use this whenever you need to add, modify, review, or debug sync integration tests, especially if the work involves MSW mocks, Linear webhooks, queue consumers, reconciliation, connection lifecycle HTTP flows, or bidirectional task/brief/initiative sync. This skill should win over generic testing advice whenever the user mentions `apps/sync`, `.integration.spec.ts`, Linear mocking, webhook tests, or sync runtime canaries.

Local testing runbook for validating Linear, Jira, and Confluence sync against the local dev stack. MUST USE this skill whenever testing sync behavior locally, debugging sync issues, verifying sync after code changes, running sync test scenarios, checking sync connections/webhooks/mappings, triggering resync, testing team switches, or verifying provider-to-Hamster state agreement. Also use when inspecting tmux logs for sync, querying sync_entity_mappings, diagnosing initiative/brief/task/blue...

Testing patterns for unit tests (Jest), integration tests, E2E tests (Playwright). Use when writing app-level tests, setting up test infrastructure, or implementing TDD workflows. For database testing (RLS, constraints, triggers, schema), use hamster-database-testing instead.

Diagnose chat thread issues from a thread_id and a user-provided symptom. Use when the user reports a misbehaving thread (missing diff preview, duplicate spawns, hallucinated failures, child threads not opening, missing Accept buttons, etc.). Walks the parent + child threads, tool_calls, participants, and relevant code paths to produce a root-cause summary and a concrete fix plan.

Horde provisioner patterns for Helm SDK rendering, multi-tenant K8s deployment, per-org isolation, and Skaffold integration

Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-have-adhd; stays on until \\\"stop adhd mode\\\".

Survey a codebase's animation and motion code as a senior motion advisor, then produce a prioritized audit and self-contained implementation plans for other agents (or cheaper models) to execute. Read-only on source code — it plans improvements, it does not apply them. Use when the user asks to \\\"improve the animations\\\", \\\"audit the motion\\\", \\\"make this app feel better\\\", or wants a roadmap of animation fixes rather than a review of a single diff.

Guides linking initiatives to goals (many-to-many with weights), assigning owners at creation, setting focus (now/next/later), enforcing 1-2 week scoping, and running discovery against a goal to spawn initiatives. Ensures every initiative has at least one linked goal — no unanchored work.

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Maintain an open pull request in a fast-moving repository. Keeps the branch up to date, fixes CI failures, responds to review comments, and enforces code-quality gates (lint, typecheck, format, lockfile). Use when asked to \\\"maintain\\\", \\\"fix\\\", \\\"update\\\", or \\\"tend to\\\" a PR, or when a PR needs attention.

Turn market signals (competitive moves, customer churn, support patterns, search console data, product analytics) into structured direction inputs. Enforces signal-noise separation, confidence levels, and routing through discovery — not delivery. Produces a ranked signal log that feeds Phase 3 (Run Discovery) of the Hamster Direction workflow.

Guides binding OGSM measures to external platforms (PostHog, Google Search Console). Enforces leading vs lagging tagging, validates the bound event/property exists before claiming live sync, and prevents stale bindings from producing stale numbers. Use when binding a measure to PostHog or Google Search Console, setting up live sync for a measure, or connecting a goal metric to an external platform.

Best practices for when to create Mermaid diagrams and how to write good ones — layout direction, readability, shape choice, and anti-patterns. Use when writing or reviewing blueprints, initiatives, or briefs that need visual structure, when deciding whether a diagram is the right format, or when a Mermaid diagram is already in a doc and needs review.

For goal-based routines that keep the Hamster method/skill/routine content library complete. Verifies the tracking table in the Longform Skills, Methods & Routine Content blueprint is current, adds missing rows, and processes gaps.

NestJS best practices and architecture patterns aligned with the Hamster Studio stack (nestjs-zod, Supabase, Vitest, BullMQ). Use when writing, reviewing, or refactoring NestJS code. Complements the hamster-nest-api skill which covers project-specific templates.

Write and update pull-request titles and descriptions for technical work. Use whenever creating, revising, or reviewing a PR body, release-style PR summary, or reviewer-facing change explanation—even if the user only says “open a PR,” “update the PR,” “make this understandable,” or provides technical notes. Put people, user outcomes, and why first; move implementation detail and validation after the plain-language explanation.

Pick the right library for a given frontend task from a curated, opinionated list — numbers, OTP inputs, charts, command menus, virtualization, drag and drop, toasts, state, styling, and more. Only runs when explicitly invoked; it does not trigger on its own.

Guides users through designing, launching, and reading A/B tests via PostHog from Hamster. Covers experiment setup, feature flag wiring, metrics, cohorts, and results analysis.

Standardizes creating and debugging PostHog insights for Hamster goal measure bindings. Covers TrendsQuery vs HogQL, the measure binding contract (aggregated_value), math types, formulas, HogQL pitfalls, and the Composio API actions needed. Use when binding a PostHog metric to a goal, debugging a measure showing 0/null, or converting HogQL to TrendsQuery.

Produce a reviewer's orientation guide for a large or foundational pull request — a structured walkthrough that explains the mental model, traces the data/code flow, recommends a reading order, and tells the reviewer how to see and test the change locally. NOT a code review and does NOT produce inline review comments. Use when the user says things like \\\"explain this PR to me\\\", \\\"walk me through this PR\\\", \\\"I was asked to review this huge PR\\\", \\\"help me understand this PR\\\"...

Build multiple genuinely different versions of a UI piece you describe, rendered behind a visual picker so you can flip through them live and promote the one that feels right. Only runs when explicitly invoked; it does not trigger on its own.

Keeps the Cloudflare Worker (worker.js) and Node.js dev proxy (dev-proxy.mjs) in sync. MUST activate when editing either proxy file.

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

Redis expert for caching, pub/sub, data structures, and distributed systems patternsUse when \\\"redis, caching strategy, cache invalidation, pub/sub, rate limiting, distributed lock, session storage, leaderboard, message queue, upstash, redis, caching, pub-sub, session, rate-limiting, distributed-lock, upstash, elasticache, memorystore\\\" mentioned.

Build custom branded animated graphics (three.js/canvas/particle effects, hero animations, generative visuals) from visual references or Figma sketches — the reference → prompt → exact-geometry → screenshot-iterate → parameterize → embed workflow. Use when the user wants an animation \\\"like X's site\\\" (Stripe, Linear, etc.), wants to turn a Figma sketch or SVG into an animated graphic, needs to embed generated animation code into Webflow/React/Vue, or wants to scale one effect across many...

Hamster team skill: refreshing a blueprint when product decisions shift.

Generate user-facing release notes for Hamster Studio. Analyzes commits and code changes between versions to produce clear, technical release notes.

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

Reviews animation and motion code against a high craft bar derived from Emil Kowalski's design engineering philosophy. Default to flagging; approval is earned.