# AI use cases for Ever Room (NxcoreAI/EverRoom) Repo: [NxcoreAI/EverRoom](https://github.com/NxcoreAI/EverRoom) | Use cases: 4 | Page: https://dedupedai.com/use-cases/nxcoreai-everroom/ EverRoom is a local-first personal workspace, built as an Electron desktop app with a local Gateway, that links files, conversations, memory, and AI agents into scoped Context Rooms. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run the Ever Room desktop workspace on a Mac: https://dedupedai.com/use-cases/nxcoreai-everroom/run-everroom-desktop-locally/ - Organize one project into a Context Room: https://dedupedai.com/use-cases/nxcoreai-everroom/build-context-room-for-project/ - Have an Agent draft a Context Doc that you review: https://dedupedai.com/use-cases/nxcoreai-everroom/draft-context-doc-with-agent-review/ - Switch Ever Room's Agent to a real model provider: https://dedupedai.com/use-cases/nxcoreai-everroom/connect-real-agent-model-to-everroom/ ## How to run the Ever Room desktop workspace on a Mac with Ever Room Repo: [NxcoreAI/EverRoom](https://github.com/NxcoreAI/EverRoom) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Clone Ever Room and start the Electron desktop app with the isolated development runtime. **Who it is for:** A developer or power user on a Mac who wants to try a local AI context workspace before connecting real sources. **The problem:** Setting up a local AI workspace often means wiring services before anything runs. The user needs a quick way to confirm the app and its Gateway start. **The outcome:** The desktop workspace opens, and the sidebar shows the Gateway state and process ID. ### Steps 1. Install Node.js 22 or newer and pnpm 11.15.1 or a compatible pnpm 11 release. 2. Run `git clone https://github.com/NxcoreAI/Everroom.git` 3. Run `cd Everroom` 4. Run `pnpm install` 5. Run `pnpm dev`, then check the sidebar for the current Gateway state and process ID. ### Prompt to paste into your agent ```text Using the EverRoom repository at , follow the Quick start in README.md to install dependencies and run pnpm dev. Report whether Electron starts and what Gateway state the sidebar shows. Verify that the Gateway runs on a loopback port as the README describes. ``` ### Caveat The README says the app is under active development and APIs, storage schemas, and agent contracts may change, so the first run may differ from later builds. 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/nxcoreai-everroom/run-everroom-desktop-locally/ ## How to organize one project into a Context Room with Ever Room Repo: [NxcoreAI/EverRoom](https://github.com/NxcoreAI/EverRoom) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Gather a project's sources, decisions, and documents into one bounded Room for scoped agent work. **Who it is for:** A project lead or developer whose files, decisions, and notes for one project are spread across folders and chats. **The problem:** Asking a general chat tool about a project pulls in unrelated material or misses the decisions that matter. The user needs one bounded place for the project's context. **The outcome:** A Context Room holds the project's linked sources, a Room wiki, and Context Docs, and an Agent session works only from that Room. ### Steps 1. Run `pnpm dev` from the Everroom folder to open the desktop workspace. 2. Create a Room for the project in Rooms. 3. Link the project's files and repositories as sources from Files or Sources. 4. Write the decisions and goal in a Context Doc inside the Room. 5. Start an Agent session from the Room and ask it a question about the project's current status. ### Prompt to paste into your agent ```text Using the EverRoom repository at , read README.md and list the steps needed to set up a Context Room for the project . Output a numbered checklist that names the Room, sources, and Context Docs to create. Verify that each step uses a feature the README lists as available. ``` ### Caveat The README marks Knowledge Rooms and Context Docs as available but still evolving, so some Room behavior may change. 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/nxcoreai-everroom/build-context-room-for-project/ ## How to have an Agent draft a Context Doc that you review with Ever Room Repo: [NxcoreAI/EverRoom](https://github.com/NxcoreAI/EverRoom) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Let a scoped Agent propose edits to a versioned document, then review the operations before they stand. **Who it is for:** A team lead or writer who wants an agent to draft a project brief but keep final ownership of the text. **The problem:** Agents that write directly into documents make it hard to see what changed. The user needs proposed edits to stay reviewable. **The outcome:** A versioned Context Doc contains an agent-drafted section, and the user has reviewed each change as an operation. ### Steps 1. Run `pnpm dev` and open a Room that contains the project sources. 2. Create a Context Doc for the brief in Documents. 3. Start an Agent session scoped to the Room and ask it to draft one section of the brief. 4. Review the proposed edits as document operations before accepting them. 5. Keep the committed version as the record and ask the Agent for a revision if needed. ### Prompt to paste into your agent ```text Using the EverRoom repository at , read the Context Docs and Agent sections of README.md. Produce a short procedure for asking an agent to draft a section of and reviewing its edits. Verify that the procedure says edits are reviewable and committed through the document kernel. ``` ### Caveat The README describes reviewable Agent edits as evolving, so review controls may differ from the description in later versions. 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/nxcoreai-everroom/draft-context-doc-with-agent-review/ ## How to switch Ever Room's Agent to a real model provider with Ever Room Repo: [NxcoreAI/EverRoom](https://github.com/NxcoreAI/EverRoom) | Difficulty: Advanced | Time: 30 minutes | Basis: From the README Set the Pi runtime and provider variables so the Agent uses a real model instead of the fake runtime. **Who it is for:** A developer who has an OpenAI-style API key and wants the workspace Agent to produce real answers instead of fallback output. **The problem:** The default fake runtime lets the app open without credentials, but it cannot test real agent work. The user needs the Pi runtime connected to a model. **The outcome:** The desktop starts with the Pi runtime, and Agent sessions call the configured model. ### Steps 1. Add the NXCORE_AGENT_RUNTIME, NXCORE_AI_PROVIDER, NXCORE_AI_MODEL, NXCORE_AI_BASE_URL, NXCORE_AI_API, and NXCORE_AI_API_KEY values from the README dotenv block to the development environment. (inferred, check the README) 2. Put the API key in your environment or a local secret store, not in a committed file. (inferred, check the README) 3. Run `pnpm dev` and start an Agent session from a Room. 4. Send a short test prompt and confirm the reply comes from the configured model rather than the fake runtime. ### Prompt to paste into your agent ```text Using the EverRoom repository at , read the Run with a real Agent model section of README.md and list the environment variables needed for the Pi runtime. Output them as a dotenv template with empty values for secrets. Verify that no real API key is written into the output. ``` ### Caveat The README says the default runtime is fake so the app opens without credentials, and Pi runtime behavior is still evolving. Risk, credentials: The Pi runtime needs an API key for your model provider, so store it outside committed files and rotate it if it leaks. 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/nxcoreai-everroom/connect-real-agent-model-to-everroom/