Claude Skills

Use during planning to discover and document test patterns and conventions for writing consistent tests.

Use during planning to discover and document available project commands (make targets, npm scripts, etc.) for success criteria.

Use BEFORE using the Task tool. Ensures you select the correct specialized subagent_type instead of generic agents.

Use BEFORE reading multiple files to understand code. Reminds you to use codebase-analyzer agent instead of manual file reading.

This skill should be used when the user asks to "implement using TDD", "test-driven development", "RED-GREEN-REFACTOR", or "write failing test first". Enforces test-first approach with RED-GREEN-REFACTOR cycle and execution-based verification.

Design developer experiences that people love. Feedback loops, cognitive load, flow state, and ADHD-friendly patterns for tools and workflows.

Answers questions about Claude Code features, configuration, and usage from local documentation synced from code.claude.com. Use when users ask about hooks, plugins, skills, MCP servers, slash commands, sub-agents, settings, permissions, sandboxing, CLAUDE.md memory files, model selection, costs, IDE integrations (VS Code, JetBrains), CI/CD (GitHub Actions, GitLab), or cloud providers (Bedrock, Vertex, Azure).

Audit para-obsidian plugin for convention drift across skills, commands, hooks, and plugin registration.

Generate AI-assisted frontmatter field suggestions using para-obsidian LLM utilities. Demonstrates 3-layer architecture (constraints, prompt-builder, orchestration) for metadata extraction.

Guide progressive summarization of undistilled resources. Finds resources with distilled:false, acts as Tiago Forte to help extract key insights through the layers approach. Use when you want to deeply learn from saved content.

Create resource notes from analyzed proposals. Use when triage orchestrator routes a proposal with proposed_template=resource. Handles frontmatter setup and original file cleanup via para_create.

Expert guidance for creating high-quality Claude Code skills following Anthropic's official best practices. Use when creating, improving, or auditing skill files.

Automatically fix SQL performance issues with step-by-step measurement. Rewrites problematic SQL patterns (functions on columns, implicit conversions), creates indexes, measures their impact, and rolls back ineffective indexes. Reports improvements at each step with cost reduction percentages.

Analyze SQL files for performance issues like full table scans, ineffective joins, and inefficient queries. Reports findings in a CI-friendly format without modifying files. Detects critical issues (FullTableScan, IneffectiveJoin) and warnings (IneffectiveSort, TemporaryTableGrouping).

Generate parameter bindings for SQL files by analyzing database schema and extracting SQL placeholders. Queries the live database for table structures, matches placeholders to column types, and generates appropriate test values based on column patterns (IDs, dates, status enums, etc).

This skill should be used when the user asks "what plugins are available", "list data science tools", "what MCP servers can I use", "enable code intelligence", or needs to discover available plugins like serena, context7, or data access skills like wrds and lseg-data.

This skill should be used when the user asks to "use Gemini Batch API", "process documents at scale", "submit a batch job", "upload files to Gemini", or needs large-scale LLM processing. Includes production gotchas and best practices.

Git workflow expert — conventional commits, history exploration, worktree management, PR creation, squash, and safety. Activates for any git-related task including committing, branching, history, and repository analysis.