# AI use cases for Loci (IvenKooLab/loci) Repo: [IvenKooLab/loci](https://github.com/IvenKooLab/loci) | Use cases: 4 | Page: https://dedupedai.com/use-cases/ivenkoolab-loci/ loci is a Python tool that builds a local searchable knowledge base from notes and project docs, answers questions with file and section citations, and shares memory across AI tools through MCP and HTTP. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Search and ask questions across your own notes: https://dedupedai.com/use-cases/ivenkoolab-loci/search-and-ask-your-notes/ - Share one memory store across MCP-connected AI tools: https://dedupedai.com/use-cases/ivenkoolab-loci/share-memory-across-ai-coding-tools/ - Review what each coding agent remembers in a web console: https://dedupedai.com/use-cases/ivenkoolab-loci/review-what-coding-agents-remember/ - Keep notes indexed and check answers before you rely on them: https://dedupedai.com/use-cases/ivenkoolab-loci/daily-notes-workflow-with-verified-answers/ ## How to search and ask questions across your own notes with Loci Repo: [IvenKooLab/loci](https://github.com/IvenKooLab/loci) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Index a notes folder, then get search hits and cited answers from the terminal. **Who it is for:** A researcher or developer with scattered Markdown notes who needs to find an old decision fast. **The problem:** Decisions and facts are buried across many files, and plain keyword search misses them. **The outcome:** A local index that returns excerpts and answers that cite the source file and section. ### Steps 1. Run `git clone https://github.com/IvenKooLab/loci.git` and then `cd loci`. 2. Run `python -m pip install -e .` and then `cp config.example.toml config.toml`. 3. Edit config.toml: fill in [llm] and [embed], and set [[sources]].path to your notes directory. 4. Run `loci ingest` to build the index. 5. Run `loci search "deployment decisions"` and then `loci ask "What did we decide about deployment, and why?"`. ### Prompt to paste into your agent ```text Using the IvenKooLab/loci repo, configure config.toml so [[sources]].path points to , run loci ingest, then run loci ask with the question . Expected output: an answer with file > section citations. Verify that each citation names a file that exists under . ``` ### Caveat Embedding and chat requests go to the providers you configure, so only local endpoints keep the whole workflow offline. Risk, credentials: Provider API keys go in config.toml or the BRAIN_LLM_API_KEY and BRAIN_EMBED_API_KEY variables, so keep them out of the repository. 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/ivenkoolab-loci/search-and-ask-your-notes/ ## How to share one memory store across MCP-connected AI tools with Loci Repo: [IvenKooLab/loci](https://github.com/IvenKooLab/loci) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Point several MCP clients at one LOCI_HOME so they read and write the same memories. **Who it is for:** A developer who switches between AI coding tools and keeps re-explaining the same project context. **The problem:** Each tool starts without the context gathered in the others, so work gets repeated. **The outcome:** Every connected MCP client searches the same index and stores durable memories in one instance directory. ### Steps 1. Complete the setup in the quick start so the instance directory contains config.toml. 2. Add the Loci server entry to each MCP client config, with command set to loci-mcp and env LOCI_HOME set to the absolute path of the instance directory. 3. If a client cannot find loci-mcp, replace the command with the absolute path to the executable. 4. Restart the client and call brain_remember to store a memory, then brain_search to retrieve it from another client. ### Prompt to paste into your agent ```text Using the IvenKooLab/loci repo, add a loci entry to the MCP config of with LOCI_HOME set to . Expected output: the client lists brain_search and brain_remember tools. Verify that a memory stored by one client is returned by brain_search in a second client. ``` ### Caveat The README says the instance directory must contain config.toml, and every client must point at the same LOCI_HOME for the sharing to work. 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/ivenkoolab-loci/share-memory-across-ai-coding-tools/ ## How to review what each coding agent remembers in a web console with Loci Repo: [IvenKooLab/loci](https://github.com/IvenKooLab/loci) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Discover agent memories on your machine and browse them in the built-in Web console. **Who it is for:** A developer who wants to audit or clean up what their coding agents have stored as memory. **The problem:** Agent memory files are scattered across tools and hard to compare or check. **The outcome:** A list of detected harness memories with per-tool stats, browsable in the browser. ### Steps 1. Make sure the instance is set up with config.toml, as in the quick start. 2. Run `loci harness` to inspect detected agent memories. 3. Run `loci serve-http` to start the HTTP backend. 4. Open http://127.0.0.1:8765/ui in a browser and browse harness memories, file views and per-tool stats. ### Prompt to paste into your agent ```text Using the IvenKooLab/loci repo, run loci harness in the instance at and then start loci serve-http. Expected output: a list of detected agent memories and a reachable console at 127.0.0.1:8765/ui. Verify that the console shows the same memory count as the loci harness output. ``` ### Caveat By default anonymous loopback clients have admin access, so keep the console on the local machine unless you set [http] local_admin = false and configure role-based tokens. 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/ivenkoolab-loci/review-what-coding-agents-remember/ ## How to keep notes indexed and check answers before you rely on them with Loci Repo: [IvenKooLab/loci](https://github.com/IvenKooLab/loci) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Watch changing files, save quick memories, and ask questions with claim verification. **Who it is for:** A knowledge worker who writes notes all day and wants answers they can trust before acting on them. **The problem:** New notes are not searchable until reindexed, and answers can state claims without evidence. **The outcome:** Changed documents stay indexed, quick memories are searchable by tag, and answers can be verified claim by claim. ### Steps 1. Run `loci watch` in a separate terminal to keep changed documents indexed. 2. Run `loci remember "Use staging for demos" --title "Demo environment"` to save a quick memory. 3. Run `loci search "demo environment" --tag memory` to confirm the memory is found. 4. Run `loci ask "How should I prepare a demo?" --verify` to get an answer with claim-by-claim verification. ### Prompt to paste into your agent ```text Using the IvenKooLab/loci repo, run loci watch in , then save a memory with loci remember and ask using loci ask with --verify. Expected output: a cited answer with verification results. Verify that the memory saved with loci remember appears in loci search results filtered by --tag memory. ``` ### Caveat The README says that after changing the embedding model you must rebuild with Loci ingest --force, or the existing index will not match. 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/ivenkoolab-loci/daily-notes-workflow-with-verified-answers/