optimize
01Command Claude Code
A command that reads a codebase's documentation and suggests which documented tasks could become runnable commands.
Command Claude Code
A command that reads a codebase's documentation and suggests which documented tasks could become runnable commands.
Command Claude Code
A command that checks a project's security against a specified checklist and labels each item as passed, failed, or not applicable.
Hook Claude Code
Runs before the agent uses a tool for Bash, Edit and Write tool calls, executing block-dangerous.py via python3. From claude-code-expert/example.
Hook Claude Code
Runs when the agent sends a notification, executing notify-slack.sh and notify-telegram.sh (2 commands). From claude-code-expert/example.
Hook Claude Code
Runs after a tool call finishes for Write and Edit tool calls, executing format-ts.sh, format-py.sh, lint-ts.sh, lint-py.sh, test-ts.sh and test-py.sh (6 commands). From claude-code-expert/example.
Settings file Claude Code
Agent settings declaring 3 hook events (PreToolUse, Notification, PostToolUse).
Skill Claude CodeCodex
A workflow skill that requires development work to move through specification, analysis, design, implementation, verification, and completion stages. It uses hooks and a state machine, which tracks and controls the current stage.
claude-code-expert/carve-harness
Agent Claude Code
An independent evaluation agent that checks generated work against completion criteria and type-safety requirements.
claude-code-expert/carve-harness
Agent Claude Code
A software-building agent that completes assigned implementation tasks in an isolated worktree, a separate copy of the code repository, including code and tests.
claude-code-expert/carve-harness
Agent Claude Code
A documentation-writing agent that creates or updates documents using confirmed build results and research findings. It covers READMEs, guides, API documentation, and architecture documents.
claude-code-expert/carve-harness
Agent Claude Code
An agent for research before implementation, using official documentation and current versions and writing the evidence and links into RESEARCH.md.
claude-code-expert/carve-harness
Agent Claude Code
An agent that creates diagrams, illustrations, and HTML mockups, which are rough visual versions of an interface.
claude-code-expert/carve-harness
Agent Claude Code
An agent that reviews whether a pull request, the proposed set of code changes, has enough tests.
claude-code-expert/carve-harness
Agent Claude Code
An agent that checks code for exposed secrets, missing authentication or permission checks, and injection risks.
claude-code-expert/carve-harness
Command Claude Code
A command that creates a commit following Conventional Commits, a shared format for writing commit messages, and pushes it to the current branch.
claude-code-expert/carve-harness
Command Claude Code
A command that commits the current changes, pulls updates, and pushes the branch; its arguments become the commit message.
claude-code-expert/carve-harness
Command Claude Code
A command that scores a fixed set of test cases, called a golden set, again and records how the scores change over time.
claude-code-expert/carve-harness
Command Claude Code
A mechanical check of an agent harness, the rules, feedback, and stored state that guide an agent's work.
claude-code-expert/carve-harness
Command Claude Code
A planning command that breaks work into units with explicit success conditions and saves the plan in the specs/ directory.
claude-code-expert/carve-harness
Command Claude Code
Audit the whole repo for over-engineering, what can be deleted.
claude-code-expert/carve-harness
Command Claude Code
Harvest ponytail: comments into a tracked debt ledger.
claude-code-expert/carve-harness
Command Claude Code
Show ponytail's measured impact scoreboard (less code, cost, time).
claude-code-expert/carve-harness
Command Claude Code
Quick reference for ponytail levels, skills, and commands.
claude-code-expert/carve-harness
Command Claude Code
Review changes for over-engineering, what can be deleted.