Agent Skills by Duy Tùng
Showing 39 vetted skills indexed across 1 GitHub repositories.
duy-tungck:journal
Analyze recent code changes to create structured technical journal entries.
duy-tungck:research
Conduct multi-source technical research with Gemini and WebSearch, producing structured reports.
duy-tungck:plan
Create detailed technical implementation plans and architecture designs.
duy-tungck:bootstrap
Automates new software project bootstrapping through AI agents and customizable workflows.
duy-tungck:security
Automate STRIDE and OWASP security audits with red-team testing and auto-fixes.
duy-tungck:test
Automate test suite execution, coverage analysis, and UI/visual tests.
duy-tungck:brainstorm
Generate and evaluate ideas through trade-off analysis in decision-making scenarios.
duy-tungck:ask
Analyze technical and architectural questions with expert-level guidance.
duy-tungck:tech-graph
Generate SVG+PNG technical diagrams from natural language descriptions.
duy-tungck:plans-kanban
Launch the ClaudeKit plans dashboard at localhost:3456 for kanban views.
duy-tungck:project-management
Track project progress and synchronize task updates with plan files.
duy-tungck:debug
Debug software issues through systematic root cause analysis and verification.
duy-tungck:worktree
Manage git worktrees for feature isolation, health audits, and cleanup.
duy-tungck:find-skills
Discover and install agent skills via the Skills CLI.
duy-tungck:context-engineering
Monitor and optimize LLM context token usage with Python scripts.
duy-tungck:scout
Locate files and gather task context using parallel agents.
duy-tungck:docs
Generate and update project documentation from codebase analysis.
duy-tungck:scenario
Generate edge cases and test scenarios from feature descriptions using 12 decomposition dimensions.
duy-tungck:xia
Extracts, compares, ports, or adapts GitHub/local features into your project via structured workflow phases.
duy-tungck:predict
Simulate a five-expert persona debate to analyze code changes for architectural, security, performance, and UX issues.
duy-tungck:skill-creator
Create and iterate on Claude skills with Python-based initialization and testing.
duy-tungck:harness
Automate scaffolding, auditing, and scoring of Claude Code harness files.
duy-tungck:coding-level
Adjust AI explanation depth and code complexity by user coding level.
duy-tungck:loop
Execute iterative optimization loops against configurable code metrics with git-based change validation.