Marielle Russo
Community@maraxen · Cambridge, MA
Graduate student at MIT researching evolution using liquid-handling robots
Agent Skills by Marielle Russo
Showing 20 vetted skills indexed across 1 GitHub repositories.
playwright-skill
Automate end-to-end browser testing with Playwright across local and remote sites.
ui-ux-pro-max
Search UI style and guideline databases with BM25 across design domains.
using-git-worktrees
Create isolated git worktrees with directory priority and gitignore safety checks.
backend-dev-guidelines
Codify layered architecture for Node.js/Express TypeScript backend projects.
test-driven-development
Guide Python teams through the red-green-refactor cycle for unit and UI tasks.
systematic-debugging
Guide engineers through a four-phase debugging process to identify root causes.
theme-factory
Apply predefined or custom themes to slides, documents, and web pages.
web-design-guidelines
Evaluate UI code against Web Interface Guidelines and output file:line findings.
prompt-engineering
Design prompt templates with few-shot and chain-of-thought patterns.
senior-architect
Plan and evaluate software architectures with diagram generation and pattern selection.
senior-fullstack
Automate fullstack project scaffolding and code-quality checks across React, Next.js, Node.js, GraphQL, and PostgreSQL.
loki-mode
Automate PRD-to-production startup development with 100+ agents.
git-pushing
Stage, commit, and push git changes with conventional commit messages.
test-fixing
Identify and fix failing tests with grouped error patterns.
writing-plans
Generate implementation plans with file paths, tests, and commands.
webapp-testing
Automate end-to-end UI testing of local web apps with Playwright in Python.
atomic-git-commit
Analyze git status and diff to group changes into atomic conventional commits.
frontend-design
Generate production-grade frontend interfaces from user specifications across React, Vue, Svelte, or HTML/CSS.
jules-remote
Dispatch atomic coding tasks to a remote AI agent with session logs.
verification-before-completion
Run verification commands and report completion only with fresh evidence.