dotfiles

Fast headless browser for QA testing and site dogfooding. (gstack)

Safety guardrails for destructive commands. (gstack)

OpenAI Codex CLI wrapper — three modes. (gstack)

Design consultation: understands your product, researches the landscape, proposes a complete design system (aesthetic, typography, color, layout, spacing, motion), and generates font+color preview... (gstack)

Designer's eye QA: finds visual inconsistency, spacing issues, hierarchy problems, AI slop patterns, and slow interactions — then fixes them. (gstack)

Post-ship documentation update. (gstack)

Restrict file edits to a specific directory for the session. (gstack)

Full safety mode: destructive command warnings + directory-scoped edits. (gstack)

Systematic debugging with root cause investigation. (gstack)

YC Office Hours — two modes. (gstack)

CEO/founder-mode plan review. (gstack)

Designer's eye plan review — interactive, like CEO and Eng review. (gstack)

Eng manager-mode plan review. (gstack)

Report-only QA testing. (gstack)

Systematically QA test a web application and fix bugs found. (gstack)

Weekly engineering retrospective. (gstack)

Pre-landing PR review. (gstack)

Import cookies from your real Chromium browser into the headless browse session. (gstack)

Set up gbrain for this coding agent: install the CLI, initialize a local PGLite or Supabase brain, register MCP, capture per-remote trust policy. (gstack)

Ship workflow: detect + merge base branch, run tests, review diff, bump VERSION, update CHANGELOG, commit, push, create PR. (gstack)

Keep gbrain current with this repo's code and refresh agent search guidance in CLAUDE.md. Wraps the gstack-gbrain-sync orchestrator with state (gstack)

Clear the freeze boundary set by /freeze, allowing edits to all directories again. (gstack)

Find "boring-monetized" niches for indie hackers — groups already paying for bad solutions — using 6 bulletproof principles (PROVEN PAYMENT, DOLLAR PAIN, REACHABLE AUDIENCE, SCHLEP MOAT, WORKFLOW LOCK-IN, 24-HOUR SALE TEST). Use when the user wants to hunt for a startup idea, find a side project to monetize, validate a niche before building, or asks "what should I build", "find me a niche", "find an idea", "indie hacker ideas", "boring business ideas", or "niche research". Not for brainstorming new features for an existing product.

Autonomous Goal-directed Iteration. Apply Karpathy's autoresearch principles to ANY task. Loops autonomously — modify, verify, keep/discard, repeat. Supports bounded iteration via Iterations: N inline config.

Steve Jobs product review. Scores the entire product 0-10, demands coherence across every surface. Use after any UI/UX/product decision. Use when asked to "product review", "jobs review", "score this product", "UX critique", or "product critique". Proactively suggest when the user has made UI/UX/product decisions that should be reviewed before shipping.

Conduct market research, competitive analysis, and industry intelligence. Use when the user wants market sizing, competitor comparisons, OSS landscape scans, distribution analysis, or research that informs build-or-skip decisions.

Formal evaluation framework implementing eval-driven development (EDD) for capability and regression testing.

Create git worktrees for parallel feature development. Use when working on multiple features simultaneously.