SilviaAre95
UserAn open-source way of work for AI-assisted building — Claude Code plugins with second-brain (Obsidian) support and tracker (Linear) integration
Categories
Indexed Skills (44)
adr-writer
Write and manage Architecture Decision Records — generate an ADR for a decision, scaffold the docs/adr/ infrastructure on first use, list or summarize existing ADRs
error-handling
Design or implement error handling patterns — custom error classes, error boundaries, API error responses
pipeline-design
Design data pipelines — ETL/ELT flows, scheduling, error handling, and monitoring strategy
pipeline-verify
Verify a data pipeline in a dev environment — run it against a bounded sample and assert schema, counts, idempotency, and clean logs. The data-platform counterpart of driving a web app in a browser.
schema-review
Review database or data warehouse schemas for normalization, performance, naming, and query patterns
sql-optimizer
Analyze and optimize SQL queries — identify performance issues, suggest rewrites, and explain execution plans
design-system
Audit or scaffold a design system — tokens, component inventory, Tailwind config, and usage consistency
heuristic-eval
Evaluate a UI against Nielsen's 10 usability heuristics with severity ratings and fix recommendations
layout-review
Review page layouts for visual hierarchy, spacing, alignment, and responsive behavior across breakpoints — includes touch targets and mobile-first checks
user-flow-analysis
Map and analyze user flows through the application — identify friction points, drop-off risks, and optimization opportunities
ci-pipeline
Generate or review CI/CD pipeline configs — GitHub Actions, Railway, Vercel, or generic CI
dockerfile
Generate or review Dockerfiles with multi-stage builds, security hardening, and optimized layer caching
infra-review
Review infrastructure configs — Docker, CI/CD, deploy targets, env vars, and production readiness
accessibility-check
Audit accessibility — WCAG 2.1 AA code compliance (semantic HTML, ARIA, keyboard, contrast) plus UX experience mode evaluating real usability for people with disabilities
component-builder
Scaffold React/Next.js components with proper types, props, accessibility, and Tailwind styling
styling-review
Review Tailwind CSS usage, design consistency, and styling patterns across components
prd-writer
Generate a Product Requirements Document from a feature idea — problem, solution, scope, success metrics
task-breakdown
Break a feature or epic into implementable tasks with estimates, dependencies, and suggested order
user-stories
Generate user stories with acceptance criteria from a feature description or existing code
bug-review
Analyze code changes or bug reports to identify root cause, verify fix completeness, and check for regressions
edge-case-finder
Identify edge cases, boundary conditions, and failure modes in code that tests should cover
regression-check
Analyze code changes for potential regressions — check affected paths, downstream consumers, and test coverage gaps
dependency-check
Audit project dependencies for known vulnerabilities, outdated packages, and unnecessary bloat
iam-review
Review identity and access management — auth flows, RBAC, session management, and permission boundaries
create-skill
Generate a new SKILL.md file with proper frontmatter, structure, and $ARGUMENTS support
nextjs-vercel
Stack context for Next.js + Vercel projects — App Router, RSC, Prisma, Tailwind, and Vercel deployment conventions
python-gcp
Stack context for Python + GCP projects — FastAPI/Flask, BigQuery, Cloud Run, Pub/Sub, and GCP conventions
terraform
Stack context for Terraform/IaC projects — module structure, state management, and GCP/AWS provider conventions
api-docs
Generate API documentation by reading route handlers — endpoints, request/response schemas, auth requirements
readme-gen
Generate a comprehensive README.md by analyzing the project's code, config, and structure
e2e-tests
Generate end-to-end tests that verify complete user workflows through the UI — Playwright or Cypress
integration-tests
Generate integration tests that verify component interactions — API routes, database queries, service layers
unit-tests
Generate unit tests for functions, classes, or modules — isolated, fast, with edge cases and mocks
code-audit
Security audit of application code — OWASP Top 10, injection vectors, auth flaws, data exposure
security-scan
Scan your Claude Code configuration (.claude/ directory) for security vulnerabilities, misconfigurations, and injection risks using AgentShield. Checks CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions.
feature-bank
Prevent agent drift and silent feature mutation by enforcing a source-of-truth feature bank. Trigger this skill WHENEVER you are about to edit, create, or delete code in /src, /app, /lib, /api, /components, /pages, or similar code directories — no matter how small the change. Also trigger on requests like "add a feature", "implement X", "remove Y", "refactor Z", "change how X works", "populate the feature bank", "backfill features", or when the user describes product behavior. The skill runs a hard preflight check against /docs/features/ before any code is written, blocks scope creep via explicit non_goals, requires a diff-first approval flow for intentional spec changes, and runs a postflight to update metadata and changelog. For existing codebases with no feature bank, it runs an interactive Backfill flow to reverse-engineer a V0 from the code. Do NOT skip this skill to "move faster"; skipping is the drift it is designed to prevent.
repo-protection
Generate or review GitHub repository protection — branch rulesets, secret scanning, Dependabot, and Actions token hardening
system-design
Design system architecture for a feature or service — produces component diagrams, data flow, and tech decisions
tradeoff-analysis
Compare architectural or technical options with a structured pros/cons/recommendation analysis
api-design
Design REST or GraphQL API endpoints with request/response schemas, validation, auth, and error handling
db-schema
Design or review Prisma database schemas with relations, indexes, and migration strategy
conventions
Apply wayworks working conventions: simplicity-first, root-cause fixes, explicit error handling, conventional commits. Language-agnostic — stack specifics load via stack profiles.
expo-mobile
Stack context for Expo / React Native projects — Expo Router, secure storage, permissions, and mobile-specific conventions
generic
Generic stack context for projects without a specific stack profile — language-agnostic conventions
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.