Agent
An agent that independently reviews recent code changes against the project's requirements, design, testing approach, and security expectations. TDD means writing tests before completing the implementation cycle.
Agent
An independent completion checker that verifies whether a project's stated done conditions are actually met before release or a completion claim.
Agent
An agent that reviews newly written end-to-end tests, which test a complete user flow across the application. It compares those tests with the project's requirements and testing conventions.
Agent
An independent check of documents handed to another agent or session. It reads them without the original conversation and tests whether the next agent has enough information to start the stated task.
Agent
An independent review of the requirements and basic design documents before implementation begins. It checks whether the problem, solution, boundaries, architecture, data model, process steps, and completion checks fit together.
Skill Claude CodeCodex
A basic-design phase for deciding how a confirmed requirement should be built before implementation. It covers the system structure, data relationships, step-by-step behavior, and division of responsibilities through discussion.
Skill Claude CodeCodex
A requirements-definition step used before designing or building a requested change. It clarifies the underlying reason, desired outcome, boundaries, and affected parts of the project.
Skill Claude CodeCodex
A set of rules for working with the requester as a thinking partner while reaching a decision. It favors shared understanding and clear, complete wording over silently guessing what the requester means.
Skill Claude CodeCodex
A guide for running end-to-end tests, which check a complete user journey through the application. It covers finding the project's instructions for setup, commands, continuous integration, and handling failures.
Skill Claude CodeCodex
A guide for writing, editing, and reviewing end-to-end tests, which test a complete user journey through the application. It finds and follows the existing testing style for the part of the project being changed.
Skill Claude CodeCodex
A Git workflow skill defines how coding work is branched, synchronized, committed, merged, and pushed. Git is a system for tracking code changes, while a worktree is a separate working copy of a repository.
Skill Claude CodeCodex
A set of rules for writing documents that another agent can use without access to the original conversation. It requires the document to include the reasons, decisions, assumptions, limits, rejected options, completed checks, and likely next steps.
Skill Claude CodeCodex
An implementation skill helps an agent follow the coding rules and architecture of the part of a project it changes. Architecture means the way a codebase is divided into parts and responsibilities.
Skill Claude CodeCodex
An options skill compares several possible approaches to a design or implementation decision, showing their advantages, drawbacks, and one recommendation.
Skill Claude CodeCodex
A structured research and experiment process for answering questions that block an important decision. It starts by agreeing on the decision to support, tests possible explanations including counterevidence, and reports a conclusion that can guide the next step.
Skill Claude CodeCodex
A retrospective skill guides a review of completed work to identify improvements and record them in project rules or procedures. A retrospective is a structured look back at what happened and what should change next time.
Skill Claude CodeCodex
A TDD implementation skill breaks coding work into small cycles: write a failing test, make it pass, review and improve the code, then commit it. TDD means test-driven development, where tests guide the implementation.
Skill Claude CodeCodex
A user-story review skill checks whether each requirement is independent, valuable, small enough, and testable. A user story is a short description of something a user needs from a product.
Agent Claude Code
A review agent for checking text before it is committed or pushed to a public code repository. It looks for organization-specific names, identifiers, paths, and other details that could identify internal work.
Agent Claude Code
A review agent for designing and improving coding-agent skills and agents. It acts as an independent pair-programming partner, checking designs against shared project rules.
Hook Claude Code
Runs before the agent uses a tool for Bash tool calls, executing block-claude-headless.sh. From sei-newbear/xp-harness.
Hook Claude Code
Runs when a session starts on startup and resume, executing git-freshness.sh (2 commands). From sei-newbear/xp-harness.
Hook Claude Code
Runs when you submit a prompt, before the agent sees it, executing git-freshness.sh. From sei-newbear/xp-harness.
Settings file Claude Code
Agent settings declaring 3 hook events (PreToolUse, SessionStart, UserPromptSubmit).