Obra
Brainstorming
Forces Claude to actually think before coding, which frankly should be the default but isn't. Blocks all implementation until you've walked through context expl
Dispatching Parallel Agents
When you've got multiple independent failures across different test files or subsystems, this cuts debugging time by running separate agents in parallel instead
Executing Plans
Takes a written implementation plan and executes it step by step with proper checkpoints. Loads the plan file, reviews it critically for gaps or unclear instruc
Finishing A Development Branch
When you finish implementing a feature branch, this handles the "now what?" moment by running your test suite first, then presenting exactly four options: merge
Receiving Code Review
When you're on the receiving end of code review feedback, this skill keeps you from either blindly implementing suggestions or getting defensive. It walks you t
Remembering Conversations
This skill prevents Claude from wasting time reinventing solutions you've already discussed. When you ask "how should I implement X" or reference past conversat
Requesting Code Review
This dispatches a code reviewer subagent that gets clean context about your changes without your session history, keeping reviews focused on the actual work pro
Subagent Driven Development
Here's a workflow for executing implementation plans by dispatching fresh Claude instances per task, then running two-stage reviews on each result. You extract
Systematic Debugging
Forces you to investigate root causes before attempting any fixes through a rigid four-phase process: evidence gathering with diagnostic logging at component bo
Test Driven Development
Forces strict test-driven development by requiring tests before any implementation code. Won't let you write production code without a failing test first, and a
Using Git Worktrees
Creates isolated git worktrees with systematic safety checks and project setup automation. Follows a clear directory priority (existing .worktrees, then worktre
Using Superpowers
This is the meta-skill that makes the entire Claude Code marketplace work. It forces Claude to actually invoke skills before doing anything, even asking clarify
Verification Before Completion
This enforces a hard rule against claiming code works without running verification commands first. It catches you before saying "tests should pass" or "build lo
Writing Plans
Takes specs and generates step-by-step implementation plans with actual code snippets, exact file paths, and specific test commands. Maps out file structure fir
Writing Skills
This is test-driven development applied to writing documentation. Instead of testing code, you're testing whether agents actually follow your instructions. You
SaaS Factory
Skills like "Obra skills" are great on their own. Paired with a pre-configured AI company, they ship features.