Claude Code Code Review and Testing Mods

Hinweise, wann /compact, /clear oder Weitermachen sinnvoll ist: Kontextfüllung, Prompt-Cache, Auto-Compact und Limits

Stops gh pr merge until the head is clean: CI green, a real review, no open threads, and the Astra line where a repo needs it

Cross-model review and delegation: /mm:review and /mm:run, built on mmrun

Shows the current branch's PR on the prompt hint row (title and its state: Draft, Ready or Merged); hover the hint row to preview a card of the spec…

A review desk panel (/reviews): Claude publishes open questions, decisions, findings, write plans, animated explainers and read-throughs; you answer…

Opens a pane showing the full diff of each file changed on this branch vs its base

Review the working diff in Neovim beside Claude Code, or the last response in a pane: annotate lines, ask questions inline and prepare a follow-up…

A recap of the conversation you just cleared, as a card to share: time worked, turns, tool calls, files and lines changed, test runs and cost

コードが変わっていなければ npm test を実行せずに前回の結果を返す

A live pane for a workstreams round: each stack layer's place in the review loop, its CI, threads, agents and collisions

Guards a shared worktree: stops `git add -A` from staging files this session did not touch, previews destructive git commands, and points at the…

Every file that changes becomes a crop: Claude's edits and yours grow a farm in its own pane, tests bring sun or storms, and passing turns fill the…

The /visual-review command shows the design of the changes against a base branch as a diagram in the browser: classes and modules with their…

Ready-made subagents for focused jobs: debugger, test writer, doc writer and migrator

After every Edit, Write, NotebookEdit or sed -i, a docked pane shows that file's uncommitted hunk with [keep] and [revert] buttons; a press dismisses…

Images in the terminal: Claude shows the pictures it makes or finds inline, as real pixels in kitty and Ghostty and as a colored-block preview…

/reviews pane: open ComposioHQ PRs waiting on reviewers, with a Slack nudge per reviewer

A Haiku test user that clicks through your locally running app and reports UI/UX issues, with a live pane of its tasks, findings and failures

Daily, weekly and monthly reviews of your work in Claude Code, summarized from the prompts you typed: projects, threads, decisions and open items

A spec-driven, test-driven workflow for working with Claude: /devflow:go routes a task to a track (feature, bug, refactor, spike, chore) and sizes…

Keeps files' git line endings (CRLF/LF) after Claude's edits and shell commands, and stops prettier checks failing on a CRLF checkout

Make your own plugins and mods, and catch what a mod breaks without a word before a session loads it

A lint gate for Claude's edits: your project's rules, a baseline of existing violations, and a PreToolUse check that refuses only edits adding new…

The owner's interface to the plugin fleet: /fleet shows each owned plugin's installed and offered version, the repository's open PRs with their CI…

Warns (status line and toast) only when RAM or disk is low: free RAM, free disk and running test runners

The current branch's GitHub PR as a bar above the prompt or a panel, with one-press commit, review and CI actions

Band above the prompt with the current branch PR, review state and CI checks

Role buttons above the prompt: check or adversarially review the working tree with Codex, hand a task over to Codex, or verify with a second Claude…

A pane and toasts for background shells, monitors and subagents: status, elapsed time, latest output line, exit code

SDD capability 08: evidence-based convergence state, regressions and recorded exceptions

⏭️ When a session gets long, writes a self-contained briefing for the next task, lets you review it, and starts it in a fresh context

/banben side pane of preview servers started by headless claude -p runs (port, effort level, done or running), using local lsof/ps; blocks opening…

Say "this error", "the tests", "this file", "what you just did" or "the plan" and what it points at goes along with your prompt: the failing output…

Shows pasted text in full in the prompt box and the transcript, and previews pasted images above the prompt and under the sent message

File tree pane that follows Claude's reads, edits and searches, with git status, search, preview and path references

Runs your test suite after every turn in which Claude changed source, and keeps the result above the prompt: green with the pass count, red with the…

Stops Claude from starting a heavy job (typecheck, tests, builds) while another one is already running on the machine

Preview images pasted into the prompt box, above the prompt, before you send them, and annotate them on macOS

/mod-doctor checks every installed mod in one report: hook scripts, plugin mods loaded, Node, Claude Code version, errors and ESLint timings in the…

Routes bare pytest through the project interpreter and requires a test file or node-id per pytest run

Three daily quests from the repo's TODOs, open issues and failing tests, in a pane, with a toast for each one cleared

Keeps review findings (H1, M3, L2...) per repo across sessions and ticks them off when their PR merges

Denies a Bash merge (gh pr merge, git merge on main, push onto main) unless the latest human message says the word merge; ship, push and deploy do…

A fast second opinion on every edit: a small model flags removed error handling, deleted tests, secrets and more

/mods opens a panel beside the chat with the mods you made and the plugins you installed: on or off, tests, updates you can apply, and your repo's…

Live progress bands above the prompt for long-running work Claude starts: tests, builds, scripts, logs

Documentation: leave a trail a reviewer can follow, and keep what can't be rebuilt

A list of the pull requests this session created or edited, with CI and review state, shown in a pane; watches CI after a push, tells the agent about…