# AI use cases for Vomit (zachahn/vomit) Repo: [zachahn/vomit](https://github.com/zachahn/vomit) | Use cases: 3 | Page: https://dedupedai.com/use-cases/zachahn-vomit/ Vomit is a Go tool that rewrites Claude Code's verbose output into plain English by sending it through a local LLM. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Shorten Claude Code output with a local LLM: https://dedupedai.com/use-cases/zachahn-vomit/shorten-claude-code-output-with-local-llm/ - Follow a Claude session with translated output: https://dedupedai.com/use-cases/zachahn-vomit/follow-claude-sessions-side-by-side/ - Point Vomit at a different local LLM server: https://dedupedai.com/use-cases/zachahn-vomit/swap-local-llm-backend-for-vomit/ ## How to shorten Claude Code output with a local LLM with Vomit Repo: [zachahn/vomit](https://github.com/zachahn/vomit) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Hook Vomit into Claude Code so a local model rewrites its verbose replies in plain English. **Who it is for:** A developer using Claude Code who wants to save tokens and read shorter replies, and has a local LLM server running. **The problem:** Claude Code's output is often long and token-heavy, and the developer wants it condensed without sending it to another hosted service. **The outcome:** Claude Code's replies in the terminal are rewritten by a local model, and the original output is still available with ctrl-o. ### Steps 1. Install Llama.app from https://github.com/ggml-org/Llama-macOS, then download GPT-OSS 20B through it (the README's recommended model). 2. Run `go install github.com/zachahn/vomit@latest`. 3. Run `vomit init` to set up the connection details to your LLM. 4. Run `vomit scrub -claude` and follow its instructions on replacing Claude's output via hooks. 5. Start Claude Code and send a prompt. Press ctrl-o to compare with Claude's original output. ### Prompt to paste into your agent ```text Install the zachahn/vomit repo with `go install github.com/zachahn/vomit@latest`, then run `vomit init` against the local LLM server at and follow the output of `vomit scrub -claude` to wire up the Claude Code hook. Report the setup steps you completed, and verify by sending one Claude Code prompt and confirming the reply is rewritten, with the original visible after ctrl-o. ``` ### Caveat The README says the local model can hallucinate and the tool is slow, and that it was only tested on Mac. 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/zachahn-vomit/shorten-claude-code-output-with-local-llm/ ## How to follow a Claude session with translated output with Vomit Repo: [zachahn/vomit](https://github.com/zachahn/vomit) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Translate an existing Claude session's tokens from the side without installing the hook. **Who it is for:** A developer who wants to read a Claude session in English without changing how Claude Code is configured. **The problem:** The developer does not want to modify Claude Code's hooks, but still wants the session's output rewritten for easier reading. **The outcome:** The developer sees a translated stream of Claude's output for a chosen session, or for the latest one. ### Steps 1. Install Llama.app or Ollama and get a local model running, as described in the README. 2. Run `go install github.com/zachahn/vomit@latest`. 3. Run `vomit list` to see Claude session identifiers. 4. Run `vomit tail [ ]` with a session identifier from the list, or with no identifier to follow the latest session. ### Prompt to paste into your agent ```text Using the zachahn/vomit repo, run `vomit list` to show Claude session identifiers, then run `vomit tail` for the session , or for the latest session if none is given. Report the session ID you followed and confirm that translated output is printed. Verify that the output reads as plain English rather than raw Claude tokens. ``` ### Caveat The README marks this non-invasive mode as deprecated, so it may change or be removed. 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/zachahn-vomit/follow-claude-sessions-side-by-side/ ## How to point Vomit at a different local LLM server with Vomit Repo: [zachahn/vomit](https://github.com/zachahn/vomit) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Use Ollama, Apfel, or another OpenAI-API server as the model behind Vomit. **Who it is for:** A developer who already runs a local model server and does not want to install Llama.app. **The problem:** The developer already has a local model server and wants Vomit to use it instead of a new one. **The outcome:** Vomit sends Claude's output to the chosen local server and returns rewritten text. ### Steps 1. Install and start a server from the README list, such as Ollama (https://github.com/ollama/ollama) or Apfel (https://github.com/Arthur-Ficial/apfel). 2. Run `go install github.com/zachahn/vomit@latest`. 3. Run `vomit init` and enter the connection details for the server you started. 4. Run `vomit scrub -claude` and follow its hook instructions. 5. Send a Claude Code prompt and confirm the reply is rewritten. ### Prompt to paste into your agent ```text Install zachahn/vomit with `go install github.com/zachahn/vomit@latest`, then run `vomit init` pointed at the local server that is already running. Report which server and connection details you entered, and verify by checking that a Claude Code reply is rewritten after `vomit scrub -claude` is applied. ``` ### Caveat The README says Apfel is not very good, and it does not confirm that every OpenAI-API server works, since that line ends in a question. 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/zachahn-vomit/swap-local-llm-backend-for-vomit/