Niels Madan
Community@nielsmadan · Berlin, Germany
Agent Skills by Niels Madan
Showing 27 vetted skills indexed across 1 GitHub repositories.
rn-upgrade
Automates React Native upgrades including dependency compatibility analysis and native code changes for Android and iOS.
read-docs
Search internal project documentation by keywords and prioritize Markdown content matches.
review-plan
Coordinate parallel specialized agents to review software implementation plans for issues.
explain
Explain unfamiliar code grouped by logical concepts with optional syntax details.
perf-test
Set up, execute, and analyze performance tests for code, endpoints, or services.
theme-factory
Applies visual themes to presentations, documents, reports, and HTML pages.
test
Review and generate unit tests for code using Jest, pytest, or Go testing.
debug-log
Insert debug logging statements to trace code execution and variable states.
code-review
Review code changes for bugs, architecture, security, and best practices.
review-history
Analyze Git history to identify code changes and regressions.
Extracts text and structured data from PDF documents using Python libraries.
todo
Creates prioritized Todoist tasks and selects pending ones via CLI and Git remote origin mapping.
review-perf
Analyze code for performance bottlenecks and provide severity-ranked fixes.
review-comments
Review and clean up low-quality code comments across staged, changed, or all files.
skill-creator
Guide creation and structuring of AI skills with SKILL.md best practices.
optimize-seo
Audit web pages and generate SEO meta tags, structured data, and image attributes.
flutter-upgrade
Upgrade Flutter SDK and project dependencies to specified stable versions.
research-online
Research programming topics by spawning parallel sub-agents across documentation, GitHub, Stack Overflow, and web searches.
workbench
Execute Python, Node.js, Go, and Bash scripts in a Docker-sandboxed environment.
resolve-conflicts
Identify conflicted files and provide Git commands to resolve them.
doc
Review and generate code documentation against defined principles and templates.
review-security
Audit code for vulnerabilities, hardcoded secrets, and insecure patterns.
cld-md-improver
Audit CLAUDE.md files for quality and suggest improvements.
review-logs
Analyze Claude Code session transcripts to identify failure patterns and suggest fixes.