Jixiang Ding
Community@sylvanding · Beijing, China
PhD Candidate in Computer Science at BIT | AI Infra, DL&ML, AGI | @zingspark | @Fullive-AI
Agent Skills by Jixiang Ding
Showing 20 vetted skills indexed across 1 GitHub repositories.
frontend-design
Generate React 18, TypeScript, and TailwindCSS interfaces for scientific web applications.
agent-browser
Automate web navigation, form filling, scraping, and screenshots via headless browser CLI.
file-todos
Create standardized markdown todo files with YAML frontmatter from code review feedback.
proof
Create and share markdown documents via Proof's web API with inline collaboration.
agent-native-architecture
Design agent-native architectures with agents as first-class citizens.
compound-docs
Capture verified technical solutions into categorized markdown docs with validated YAML frontmatter.
rclone
Transfer and sync files across multiple cloud storage providers via a single command-line interface.
orchestrating-swarms
Coordinate multiple autonomous AI agents for multi-step software engineering workflows.
andrew-kane-gem-writer
Create Ruby gems with Andrew Kane's architecture and Rails integration patterns.
every-style-editor
Review written content for compliance with Every's style guide.
setup
Automates Codeforces and other programming contest support.
document-review
Review brainstorm and plan documents for clarity gaps and scope risks.
skill-creator
Guide custom skill creation with validation and zip packaging.
create-agent-skills
Create Claude Code skills with YAML frontmatter and XML structure.
brainstorming
Clarify feature requirements and explore implementation options before technical planning.
git-worktree
Create, list, switch, and clean up isolated Git worktrees with confirmations.
dspy-ruby
Build type-safe LLM applications in Ruby with Sorbet signatures.
gemini-imagegen
Generate, edit, and compose images via the Gemini API.
dhh-rails-style
Enforce DHH and 37signals Ruby on Rails coding conventions.
resolve-pr-parallel
Resolve GitHub pull request review comments via parallel agent processing.