productivity▌

You are a browser automation expert who has debugged thousands of flaky tests and built scrapers that run for years without breaking. You've seen the evolution from Selenium to Puppeteer to Playwright and understand exactly when each tool shines.

You are an expert code reviewer conducting a thorough evaluation of local uncommitted changes. Your review must be structured, systematic, and provide actionable feedback including improvement suggestions.

PREREQUISITE: Read ../gws-shared/SKILL.md for auth, global flags, and security rules. If missing, run gws generate-skills to create it.

Manage multiple local CLI agents in parallel tmux sessions with task assignment and monitoring. \n \n Start, stop, and monitor agents running in isolated tmux sessions with log tailing and health checks \n Assign tasks to specific agents and track their execution output in real time \n Schedule recurring agent work via cron integration for automated workflows \n Requires tmux and Python 3; agents are configured in a local agents/ directory \n

Technical rigor in code review: receiving feedback critically, requesting systematic reviews, and verifying before any completion claims. \n \n Three core practices: receiving feedback with technical evaluation over performative agreement, requesting reviews via code-reviewer subagent, and verification gates requiring evidence before status claims \n Includes decision tree and protocols for handling feedback from human partners versus external reviewers, with explicit rules against performative

Autonomous AI pentester that analyzes source code, identifies vulnerabilities, and executes real exploits to prove security flaws. \n \n Covers OWASP Top 10 categories: injection, XSS, SSRF, broken authentication, and broken authorization with 96.15% exploit success rate on security benchmarks \n Requires explicit written authorization before testing; includes mandatory safety checks to prevent accidental targeting of production systems \n Runs in Docker with integrated security tools (Nmap, Sub

Memory architecture for agents: retrieval strategies that determine whether agents remember or forget. \n \n Covers five memory types: short-term (context window), long-term (vector stores), working memory, episodic memory, and semantic memory, each suited to different information patterns \n Emphasizes retrieval as the core challenge; provides chunking strategies, embedding quality guidance, and metadata filtering to surface the right memories at decision time \n Includes anti-patterns like sto

Structured canvas for evaluating AI product ideas across outcomes, hypotheses, risks, and positioning. \n \n Synthesizes 10 strategic components: business outcomes, customer outcomes, problem framing, solution hypotheses, positioning, assumptions, PESTEL risks, value justification, success metrics, and next steps \n Designed for AI-specific uncertainty; treats solutions as testable bets rather than commitments, with lightweight \"Tiny Acts of Discovery\" experiments built in \n Outcome-driven fr

Build self-serve acquisition and expansion motions. But first, figure out if PLG is even the right motion for your product.

Initialize Subframe in a project to render pixel-perfect designed pages in your local development environment. \n \n Detects project type (new, existing non-Subframe, or already setup) and follows the appropriate setup path with CLI scaffolding or integration \n Configures Tailwind CSS (v3 or v4), import aliases, and syncs all Subframe components to a designated directory \n Requires manual font configuration via Google Fonts links based on your project's theme \n Supports Vite, Next.js, and Ast