San-Hsien
Community@SanHsien · Taiwan
Personal & open-source projects · local-first tools · verifiable AI workflows · Windows-first software
Agent Skills by San-Hsien
Showing 25 vetted skills indexed across 1 GitHub repositories.
SanHsienidea-refine
Refines raw ideas into actionable one-pagers through structured divergent and convergent thinking.
SanHsiendebugging-and-error-recovery
Diagnose test failures, build errors, and runtime bugs through structured root-cause triage.
SanHsienusing-agent-skills
Routes incoming tasks to the appropriate engineering workflow skill by development phase.
SanHsientest-driven-development
Implements code changes using the red-green-refactor test-driven development loop.
SanHsienplanning-and-task-breakdown
Decomposes specs into ordered, verifiable implementation tasks with acceptance criteria.
SanHsienincremental-implementation
Implements code changes in thin, tested, individually committed vertical slices.
SanHsienapi-and-interface-design
Designs stable REST APIs, TypeScript contracts, and idempotent endpoints with consistent error semantics.
SanHsienci-cd-and-automation
Configures CI/CD pipelines with quality gates, deployment strategies, and rollback workflows.
SanHsiencontext-engineering
Curates rules files, specs, and source context for AI coding agents.
SanHsienfrontend-ui-engineering
Builds accessible, responsive, production-quality user interfaces following design system standards.
SanHsienshipping-and-launch
Guides production launches with pre-launch checklists, staged rollouts, monitoring, and rollback plans.
SanHsienspec-driven-development
Creates structured specifications with gated review phases before writing code.
SanHsienconstraint-driven-development
Defines project quality gates in CONSTRAINTS.md and detects diff changes that weaken them.
SanHsiencode-review-and-quality
Reviews code changes across correctness, readability, architecture, security, and performance before merge.
SanHsiensecurity-and-hardening
Hardens web application code against OWASP vulnerabilities through threat modeling and validation patterns.
SanHsiendeprecation-and-migration
Plans deprecation and incremental migration of legacy systems, APIs, and database schemas.
SanHsienobservability-and-instrumentation
Instruments production code with structured logs, metrics, traces, and symptom-based alerts.
SanHsieninterview-me
Extracts underlying user intent through one-question-at-a-time interviews before planning or coding.
SanHsiencode-simplification
Refactors code to reduce complexity while preserving exact behavior and passing existing tests.
SanHsiendoubt-driven-development
Subjects non-trivial decisions to fresh-context adversarial review before they stand.
SanHsiensource-driven-development
Verifies framework code against official documentation and cites sources for every implementation decision.
SanHsienperformance-optimization
Diagnose and fix frontend, backend, and database performance bottlenecks through measurement-driven optimization.
SanHsienbrowser-testing-with-devtools
Tests and debugs web pages in real browsers via Chrome DevTools MCP.
SanHsiengit-workflow-and-versioning
Structures git commits, branching, and semantic versioning into disciplined workflows.
Frequently Asked Questions About San-Hsien
FAQPage SchemaWhat tasks can I accomplish with San-Hsien's skills?▼
The 25 skills cover the full delivery lifecycle: spec drafting, task breakdown, test-driven development, root-cause debugging, multi-axis code review, OWASP security hardening, performance optimization, observability instrumentation, CI/CD setup, git versioning, and production launch checklists with rollback strategy.
Who are these skills designed for?▼
Software engineers and coding agents who want disciplined, verifiable development practices. They suit developers handling ambiguous requirements, high-stakes changes like production auth or irreversible migrations, and teams needing enforced quality bars, accessibility standards, and documented architecture decisions.
How do the skills get invoked in practice?▼
The using-agent-skills meta-skill governs discovery and invocation at session start, matching each piece of work to the right skill. Trigger phrases like 'ideate', 'stress-test my plan', or 'interview me' activate specific skills; browser-testing-with-devtools additionally requires the chrome-devtools MCP server configured.
Are San-Hsien's skills free and open source?▼
Yes. The organization profile describes personal and open-source projects with a local-first, verifiable focus across 52 public repositories. The skills are published openly in the registry with no indicated licensing fees or commercial tiers.
What prerequisites or dependencies do these skills require?▼
Most skills are methodology-driven and need no external dependencies. The exception is browser-testing-with-devtools, which requires the chrome-devtools MCP server to be configured for DOM inspection, console error capture, network analysis, and performance profiling in real browsers.