# AI use cases for Claude Code Mods (hamzafer/claude-code-mods) Repo: [hamzafer/claude-code-mods](https://github.com/hamzafer/claude-code-mods) | Use cases: 5 | Page: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/ A set of Claude Code plugins, built on hooks, that add live status lines above the prompt plus guards, panes and games that you can install one at a time. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Stop Claude from running destructive commands unseen: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/guard-repo-from-destructive-commands/ - Watch Codex and subagent code reviews run live: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/watch-code-reviews-live/ - Keep context fill and plan usage above the prompt: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/keep-context-and-plan-usage-in-view/ - Preview Markdown edits as GitHub renders them: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/preview-markdown-edits-like-github/ - Track OpenAI API credit and spend above the prompt: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/track-openai-api-credit-in-status-bar/ ## How to stop Claude from running destructive commands unseen with Claude Code Mods Repo: [hamzafer/claude-code-mods](https://github.com/hamzafer/claude-code-mods) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Hold rm -r, force pushes and migrations, and see what they would delete before they run. **Who it is for:** A developer who lets Claude Code run shell commands and wants a check before any delete, force push or migration goes through. **The problem:** A single rm -r or force push issued by an agent can remove files or rewrite history before anyone reads the command. **The outcome:** Risky commands stop at a preview that shows what they would delete, and they cancel after 60 seconds with no answer. ### Steps 1. Run `npx claude-code-mods@latest`, tick blast-radius in the checklist, and press Enter to install it. 2. Restart Claude Code. 3. Ask Claude Code to delete a test folder you do not need, for example with rm -r. 4. Read the preview of what would be deleted, then answer yes or no. 5. Leave the prompt unanswered once and confirm the command is cancelled after 60 seconds. ### Prompt to paste into your agent ```text In the hamzafer/claude-code-mods repo, install the blast-radius mod so Claude Code holds risky commands. Ask Claude Code to run rm -r on and confirm the expected output is a preview of the files that would be deleted, with the command cancelled if no answer comes within 60 seconds. ``` ### Caveat The installer edits the Claude profile that the claude command uses, so check the profile path shown at the top of the checklist first. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/guard-repo-from-destructive-commands/ ## How to watch Codex and subagent code reviews run live with Claude Code Mods Repo: [hamzafer/claude-code-mods](https://github.com/hamzafer/claude-code-mods) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README See each running code review as a line above the prompt, with its findings when it ends. **Who it is for:** A developer who starts Codex reviews or review subagents and wants to follow progress without switching windows. **The problem:** Code reviews run in the background, and it is hard to tell which one is still going, which model it uses, or what it has found so far. **The outcome:** Each running review shows one live line with its model, target, elapsed time and Codex's latest output, and a toast lists the findings when it ends. ### Steps 1. Run `npx claude-code-mods@latest`, tick review-watch in the checklist, and press Enter. 2. Restart Claude Code. 3. Start a code review with Codex or a review subagent in Claude Code. 4. Watch the live line above the prompt for the model, target, elapsed time and latest output. 5. When the review ends, read the toast that lists the findings. ### Prompt to paste into your agent ```text In the hamzafer/claude-code-mods repo, install review-watch and start a code review of in Claude Code. Expect one live line per running review showing model, target and elapsed time, followed by a findings toast when it finishes; verify the toast appears. ``` ### Caveat The README says review-watch works with Codex or a review subagent, so a review started some other way will not show a line. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/watch-code-reviews-live/ ## How to keep context fill and plan usage above the prompt with Claude Code Mods Repo: [hamzafer/claude-code-mods](https://github.com/hamzafer/claude-code-mods) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README Track how full your context is and how much of your plan you have used, in bars above the prompt. **Who it is for:** A developer on a Claude plan who runs long sessions and wants to know when to compact or when a usage limit is near. **The problem:** Long sessions fill the context window and plan limits reset on a schedule, and neither is visible while you type. **The outcome:** A stacked context bar and a usage meter sit above the prompt, showing fill, token counts, the 5-hour and 7-day usage and the reset countdown. ### Steps 1. Run `npx claude-code-mods@latest`. 2. Tick context-bar and usage-meter in the checklist and press Enter. 3. Restart Claude Code. 4. Type `/context-bar` to see the context window by category and where it compacts. 5. Press `o` in the checklist if you want to reorder the bands above the prompt. ### Prompt to paste into your agent ```text In the hamzafer/claude-code-mods repo, install the context-bar and usage-meter mods. Run /context-bar in Claude Code on and expect a stacked bar with token counts and the compact point; verify the usage meter shows the 5-hour and 7-day usage. ``` ### Caveat The usage meter reports plan usage as the README describes it, so its numbers depend on what Claude Code exposes on your plan. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/keep-context-and-plan-usage-in-view/ ## How to preview Markdown edits as GitHub renders them with Claude Code Mods Repo: [hamzafer/claude-code-mods](https://github.com/hamzafer/claude-code-mods) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README See the Markdown files Claude edits rendered like GitHub, with before and after side by side. **Who it is for:** A writer or developer who has Claude Code edit README files or docs and wants to check the rendered result before committing. **The problem:** Markdown edits are hard to judge in raw text, and rendering differences from GitHub show up only after a push. **The outcome:** Edited Markdown files open as a GitHub-style render with the before and after shown side by side. ### Steps 1. Run `npx claude-code-mods@latest`, tick md-preview in the checklist, and press Enter. 2. Confirm Chrome is installed and your terminal shows images, as the README requires. 3. Restart Claude Code. 4. Ask Claude Code to edit a Markdown file such as README.md. 5. Type `/md` to open the rendered preview and compare before and after. ### Prompt to paste into your agent ```text In the hamzafer/claude-code-mods repo, install md-preview, then ask Claude Code to edit . Run /md and expect a GitHub-style render with before and after side by side; verify the edited text appears in the after view. ``` ### Caveat md-preview needs Chrome and a terminal that shows images, so it will not work in a plain text-only terminal. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/preview-markdown-edits-like-github/ ## How to track OpenAI API credit and spend above the prompt with Claude Code Mods Repo: [hamzafer/claude-code-mods](https://github.com/hamzafer/claude-code-mods) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README See an estimated OpenAI balance, today's spend and the last call without leaving Claude Code. **Who it is for:** A developer who pays for OpenAI API usage and wants to watch credit drain while working with Claude Code. **The problem:** API credit can run down unnoticed, and it is not clear which model or call used most of the money. **The outcome:** A balance gauge, today's spend, the model that took most of the money and the last call appear in a band above the prompt. ### Steps 1. Get an OpenAI organization Admin key, which the README says this mod requires. 2. Run `npx claude-code-mods@latest`, tick openai-balance in the checklist, and press Enter. 3. Restart Claude Code. 4. Type `/openai-balance` to open the balance view. 5. Check that the estimated balance, today's spend and the last call match your OpenAI usage page. ### Prompt to paste into your agent ```text In the hamzafer/claude-code-mods repo, install the openai-balance mod using an OpenAI organization Admin key kept in . Run /openai-balance in Claude Code and expect an estimated balance with a gauge and today's spend; verify the figures roughly match the OpenAI dashboard. ``` ### Caveat The balance is an estimate, as the mod's name and description say, so treat it as a guide and not the figure your OpenAI account shows. Risk, credentials: This mod needs an OpenAI organization Admin key, so store the key in a secret manager and never paste it into a prompt or a shared file. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/hamzafer-claude-code-mods/track-openai-api-credit-in-status-bar/