productivity▌

Design patterns for building autonomous coding agents with tool integration, permissions, and human oversight. \n \n Core agent loop architecture with think-decide-act-observe cycle; supports multi-model selection for different task types \n Tool design patterns including file operations, code understanding, terminal execution, and browser automation with schema-based definitions \n Permission system with four levels (auto, ask-once, ask-each, never) and risk assessment for approval workflows \n

One-page facilitation tool for framing business problems, surfacing assumptions, and defining learning experiments before building. \n \n Guides teams through 8 adaptive questions (one per canvas box) covering business problem, outcomes, users, benefits, solutions, hypotheses, riskiest assumptions, and smallest experiments \n Combines assumptions from multiple boxes into testable hypotheses using the template: \"We believe [business outcome] will be achieved if [user] attains [benefit] with [sol

Build, package, and sign SwiftPM-based macOS apps entirely outside Xcode using shell scripts and templates. \n \n Includes a minimal SwiftPM project skeleton in assets/templates/bootstrap/ with Package.swift, Sources structure, and version configuration ready to rename and customize \n Provides six core packaging scripts: package_app.sh for .app bundle assembly, compile_and_run.sh for dev iteration, sign-and-notarize.sh for release signing and Apple notarization, and make_appcast.sh for Sparkle

Automate pull request creation with formatted commits and branch management. \n \n Ensures you're not on the main branch, creating a feature branch if needed \n Runs Oxfmt code formatting across all affected packages before committing \n Formats commits and PR titles using package name from package.json (e.g., \" @remotion/shapes : Add heart shape\") \n Uses the gh CLI to open the pull request automatically after pushing changes \n

Detect market regimes and apply data-driven trading strategies based on historical performance patterns. \n \n Identifies 32 active market regimes (trending up, moderate bull, mixed choppy, bearish, sideways flat, low volatility) with confidence scores up to 93% \n Recommends regime-specific strategies: zero-trade capital preservation in choppy/bearish markets, active 120–200 trades/day with risk validation in confirmed uptrends, asset selection prioritization based on momentum \n Highlights cri

Register, schedule, and execute background work on iOS using the BackgroundTasks framework, background URLSession, and background push notifications.

Trace upstream data lineage to identify sources, DAGs, and dependencies feeding a table or column. \n \n Supports tracing three target types: tables, columns, and DAGs; uses Airflow DAG source code and task inspection to find producing pipelines \n Handles SQL sources (FROM clauses), external systems (S3, Postgres, Salesforce, HTTP APIs), and file-based sources; recursively traces upstream chains \n Includes column-level tracing through direct mappings, transformations, and aggregations in DAG c

A lightweight CLI tool for comparing coding agents head-to-head on reproducible tasks. Every "which coding agent is best?" comparison runs on vibes — this tool systematizes it.

Swap a face from a source image into a target video using GPU-accelerated AI processing. The source image provides the face to swap in, and the target video receives the new face.

Test website responsiveness across viewport widths, detect layout transitions, and report where layouts break. \n \n Supports three operating modes: Standard Check (8 key breakpoints), Sweep (15 widths at 160px increments), and Targeted Range (user-specified widths at 80px increments) \n Runs 8 layout checks per viewport covering horizontal overflow, text overflow, navigation transitions, content stacking, image scaling, touch targets, whitespace balance, and CTA visibility \n Detects and report