productivity▌

Reduce voluntary and involuntary churn through cancel flows, dynamic save offers, and payment recovery. \n \n Design exit surveys and targeted save offers (discounts, pauses, downgrades) matched to cancellation reasons, typically recovering 25-35% of at-risk customers \n Implement proactive retention by tracking health signals (login frequency, feature usage, support sentiment) and intervening before customers reach the cancel page \n Set up dunning sequences with smart payment retries, escalati

Set up Convex authentication with the right provider, user management, and access control patterns. \n \n Supports multiple auth providers: Convex Auth, Clerk, WorkOS AuthKit, Auth0, and custom JWT, with provider detection from repo signals \n Guides you through choosing a provider, configuring client and backend wiring, environment variables, and convex/auth.config.ts setup \n Covers authentication checks in protected functions, optional app-level user storage, and authorization patterns for ow

Run web-only libraries in Expo apps by rendering them in webviews on native and as-is on web. \n \n Supports any React web library (recharts, react-syntax-highlighter, canvas, WebGL) without modification by wrapping it in a DOM component file with the 'use dom'; directive \n Passes serializable props and async functions from native to webview, enabling bidirectional communication between native and web contexts \n Includes webview configuration options via the dom prop: scroll control, safe area

Optimized patterns for AI agent workflows, Git integration, MCP servers, and multi-agent orchestration. \n \n Covers essential commands across Claude, Gemini, and Codex CLIs, plus keyboard shortcuts for task cancellation, history search, and plan mode toggling \n Includes session management techniques (resume, rename, aliases) and Git workflows for auto-generating commits, PRs, and managing parallel branches with worktrees \n Provides MCP server configuration guidance (Playwright, Supabase, Fire

Never Use TaskOutput \n TaskOutput floods the main context window with agent transcripts (70k+ tokens). \n Rule \n NEVER use TaskOutput tool. Use Task tool with synchronous mode instead. \n Why \n \n TaskOutput reads full agent transcript into context \n This causes mid-conversation compaction \n Defeats the purpose of agent context isolation \n \n Pattern \n # WRONG - floods context\n Task(run_in_background=true)\n TaskOutput(task_id=\"...\") // 70k tokens dumped\n \n # RIGHT - isolated context

Search and prioritize your assigned issues by age, engagement, and status. \n \n Queries the current repository for issues assigned to you using the GitHub repo context \n Surfaces issue recommendations based on age, comment count, and open/closed status to help you prioritize work \n Integrates with the list_issues command to retrieve and analyze issue data from your repository \n

Transcribe audio to text using ElevenLabs Scribe or Whisper models via inference.sh CLI. \n \n Three model options: ElevenLabs Scribe v2 (98%+ accuracy with diarization), Fast Whisper V3, and Whisper V3 Large for varying speed/accuracy tradeoffs \n Supports 99+ languages, optional timestamps, speaker diarization, and translation to English \n Common workflows include meeting transcription, podcast transcripts, video subtitles, and voice note conversion \n Requires inference.sh CLI ( infsh ) inst

Strategic planning and optimization for email capture offers and gated content. \n \n Covers lead magnet types (checklists, templates, ebooks, quizzes, webinars) matched to buyer stage, with effort and timeline estimates for each format \n Provides gating strategy guidance including what to ask for on forms, how to frame the value exchange, and delivery methods (instant download, email, drip) \n Includes landing page structure, thank you page optimization, and distribution channels (blog CTAs, p

Web search with optional full-page content extraction from results. \n \n Returns real search results as JSON with optional --scrape flag to fetch complete page markdown for each result, avoiding redundant fetches \n Supports filtering by source type (web, images, news), category (GitHub, research, PDF), time range (past hour/day/week/month/year), location, and country \n Use --limit to control result count and --scrape-formats to customize output formats when extracting full content \n Part of

Analyze events through the disciplinary lens of physics, applying fundamental physical laws (conservation of energy, momentum, mass; thermodynamics; electromagnetism; relativity), quantitative modeling, dimensional analysis, and systems dynamics to understand causation, evaluate constraints, assess technological feasibility, analyze energy systems, and identify physical limits that govern complex systems.