Javier Villalba
Community@256javy · Paraguay
Just learning.
Agent Skills by Javier Villalba
Showing 44 vetted skills indexed across 1 GitHub repositories.
gstack
Automate end-to-end QA testing with a persistent headless browser.
pair-agent
Pair a remote AI agent with a browser using a setup key and scoped permissions.
benchmark
Measure browser performance metrics like TTFB, FCP, and LCP via the browse daemon.
design-html
Translate approved mockups and design plans into clean HTML/CSS markup.
plan-tune
Configure per-question preferences and inspect dual-track profiles for gstack prompts.
design-shotgun
Generate multiple AI design variants and a side-by-side comparison board for UI exploration.
plan-design-review
Rate design plans across key dimensions and suggest concrete fixes.
autoplan
Load and run CEO, design, engineering, and DX review skills sequentially.
design-consultation
Generate a design system blueprint and DESIGN.md with typography, color, layout, spacing, and motion.
learn
Consolidate and manage project learnings captured across gstack sessions.
freeze
Restrict file edits to a user-specified directory during a session.
careful
Warn before executing destructive Bash commands like rm -rf and DROP TABLE.
cso
Identifies and remediates security risks in infrastructure, software supply chains, and CI/CD pipelines.
canary
Monitor live applications for console errors and performance regressions after deployment.
open-gstack-browser
Launch an AI-controlled Chromium session with a live activity feed.
investigate
Diagnose software bugs through four structured investigation phases.
context-restore
Restore the most recent saved workspace context across branches.
document-release
Update project documentation and CHANGELOG after shipped code changes.
health
Aggregate type checking, linting, tests, dead code, and shell linting into a 0-10 health score.
gstack-upgrade
Upgrades gstack installations by detecting install type, running migrations, and summarizing changelog changes.
land-and-deploy
Merge PRs, wait for CI, and deploy to production with canary health checks.
codex
Automate guided OpenAI Codex interactions for code reviews and adversarial tests.
qa
Run tiered QA tests on web applications and commit atomic bug fixes.
qa-only
Generate a report-only QA test with health score, screenshots, and reproducible steps.