# AI use cases for LLM Wiki V2 (crow-hq/llm-wiki-v2) Repo: [crow-hq/llm-wiki-v2](https://github.com/crow-hq/llm-wiki-v2) | Use cases: 5 | Page: https://dedupedai.com/use-cases/crow-hq-llm-wiki-v2/ A Python library and web app that turns your documents into a Markdown wiki of notes and answers questions with citations to those notes, using any LLM provider. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Turn a pile of documents into a cited Markdown wiki: https://dedupedai.com/use-cases/crow-hq-llm-wiki-v2/build-a-cited-wiki-from-documents/ - Build a wiki that runs on your own computer with Ollama: https://dedupedai.com/use-cases/crow-hq-llm-wiki-v2/run-a-private-wiki-with-ollama/ - Run the wiki as a server that other tools can call: https://dedupedai.com/use-cases/crow-hq-llm-wiki-v2/serve-a-wiki-over-http-with-docker/ - Add documents to a wiki from a Python script: https://dedupedai.com/use-cases/crow-hq-llm-wiki-v2/file-documents-from-python-code/ - Find notes filed in the wrong folder and check wiki health: https://dedupedai.com/use-cases/crow-hq-llm-wiki-v2/audit-and-health-check-a-crow-wiki/ ## How to turn a pile of documents into a cited Markdown wiki with LLM Wiki V2 Repo: [crow-hq/llm-wiki-v2](https://github.com/crow-hq/llm-wiki-v2) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Drop your files into a browser page and get a wiki of Markdown notes you can ask questions of. **Who it is for:** A person with many documents in mixed formats who needs answers from them and does not want to read everything again. **The problem:** Scattered documents are hard to search and answers are hard to trace back to a source. **The outcome:** A folder of Markdown notes that answers questions and cites the notes it used. ### Steps 1. Install it with `pip install llmwiki2` (Python 3.11 or later). 2. Start it with `llmwiki2`; your browser opens. 3. Pick a provider, paste its key, press Test connection, then Save and start. 4. Drop text, Markdown, PDF, Word, PowerPoint, Excel or OpenDocument files anywhere on the page. 5. Type a question and click a cited note to read it. ### Prompt to paste into your agent ```text Using the llm-wiki-v2 repo (package llmwiki2), file the documents in into the wiki, then ask it . Expected output: the answer plus the note paths it cites. Verify that each cited note exists as a Markdown file in ~/llm-wiki. ``` ### Caveat A scanned PDF with no text layer adds nothing until you run OCR on it 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/crow-hq-llm-wiki-v2/build-a-cited-wiki-from-documents/ ## How to build a wiki that runs on your own computer with Ollama with LLM Wiki V2 Repo: [crow-hq/llm-wiki-v2](https://github.com/crow-hq/llm-wiki-v2) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Use a free local model through Ollama so documents and answers stay on your machine. **Who it is for:** A user who wants to keep documents local and avoid paying for an API key. **The problem:** Cloud LLM providers need an account and key, and some documents should not leave your computer. **The outcome:** A local wiki that answers questions using a model that runs on your computer. ### Steps 1. Download Ollama from https://ollama.com/download and install it. 2. Run `ollama pull gemma4`. 3. Install the library with `pip install llmwiki2`. 4. Run `llmwiki2`, pick Ollama on the page, press Test connection, then Save and start. 5. Drop your documents on the page and ask a question to check the cited answer. ### Prompt to paste into your agent ```text Set up the llm-wiki-v2 repo with Ollama as the provider and gemma4 as the model, then file and ask . Expected output: a cited answer. Verify that the model call went to the local Ollama server, not a remote provider. ``` ### Caveat The README says CROW's classifier is remote by default, so a fully local setup needs extra configuration or runs in classic mode without a classifier. 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/crow-hq-llm-wiki-v2/run-a-private-wiki-with-ollama/ ## How to run the wiki as a server that other tools can call with LLM Wiki V2 Repo: [crow-hq/llm-wiki-v2](https://github.com/crow-hq/llm-wiki-v2) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Start the wiki in Docker and add documents and ask questions through its HTTP API. **Who it is for:** A developer or team that wants a shared wiki service that scripts and agents can call over HTTP. **The problem:** Adding documents and asking questions by hand does not fit into an automated workflow. **The outcome:** A running wiki on port 8000 that accepts documents and questions through HTTP requests. ### Steps 1. Run `git clone https://github.com/crow-hq/llm-wiki-v2 && cd llm-wiki-v2`. 2. Run `cp .env.example .env`, then set OKF_LLM_PROVIDER and its key in the file. 3. Run `docker compose up -d --build` to start the wiki on port 8000. 4. Run `curl -X POST localhost:8000/ingest -H 'content-type: application/json' -d '{"text": "Acme refunds orders within 30 days of delivery.", "title": "Refund policy"}'` to file a note. (inferred, check the README) 5. Run `curl -X POST localhost:8000/ask -H 'content-type: application/json' -d '{"question": "How long do customers have to ask for a refund?"}'` to get a cited answer. (inferred, check the README) ### Prompt to paste into your agent ```text Clone the llm-wiki-v2 repo, start it with Docker Compose on port 8000, then POST the text in to /ingest and ask through /ask. Expected output: the ingest response and an answer with cited notes. Verify that /health reports the service as running. ``` ### Caveat With Docker the settings come from .env and the web page shows them but cannot change them. Risk, credentials: The .env file holds your provider key, so keep it out of git and out of shared logs. 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/crow-hq-llm-wiki-v2/serve-a-wiki-over-http-with-docker/ ## How to add documents to a wiki from a Python script with LLM Wiki V2 Repo: [crow-hq/llm-wiki-v2](https://github.com/crow-hq/llm-wiki-v2) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Import the Wiki class and file text or files into a wiki from your own code. **Who it is for:** A developer who has a data pipeline or script that produces documents and wants them filed into a wiki automatically. **The problem:** Documents produced by a script end up in folders with no structure, and someone has to file them by hand. **The outcome:** A Python script that files each document as a note and returns the result, including the tokens spent. ### Steps 1. Install the library with `pip install llmwiki2` (Python 3.11 or later). 2. In Python, run `from llmw2 import Wiki`, then `wiki = Wiki.from_env(bundle="./wiki")` and `wiki.init()`. 3. Set the OKF_* variables and your provider key, for example OPENROUTER_API_KEY, before creating the Wiki object. 4. Call `result = wiki.ingest(open("meeting.md").read(), title="Pricing meeting")` for each document. 5. Print `result` and open the note in ./wiki to check the filing. (inferred, check the README) ### Prompt to paste into your agent ```text Write a Python script that imports Wiki from llmw2, points it at ./wiki, and calls ingest on each file in . Expected output: one result per file with the note and token usage. Verify that each returned note appears as a Markdown file under ./wiki. ``` ### Caveat The README documents the ingest call and Wiki.from_env, but this script's exact return fields should be checked against the library guide before you rely on them. 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/crow-hq-llm-wiki-v2/file-documents-from-python-code/ ## How to find notes filed in the wrong folder and check wiki health with LLM Wiki V2 Repo: [crow-hq/llm-wiki-v2](https://github.com/crow-hq/llm-wiki-v2) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README Use the audit and health commands to review where CROW would move notes and which folders keep growing. **Who it is for:** A wiki maintainer running CROW mode who wants to review the folder layout before changing it. **The problem:** As notes pile up, some end up in folders the classifier would not choose, and folders may need splitting. **The outcome:** A list of notes the classifier would move and a summary of folders that need attention. ### Steps 1. Confirm the wiki runs in CROW mode with a classifier configured in Settings. 2. Run `llmwiki2 audit` to list notes the classifier would file in another branch. 3. Run `llmwiki2 health` to see folders that keep needing new subfolders or the LLM fallback. 4. Review the audit list, then run `llmwiki2 audit --apply` only for moves you accept. (inferred, check the README) ### Prompt to paste into your agent ```text In the llm-wiki-v2 wiki at , run llmwiki2 audit and llmwiki2 health and summarize the proposed moves and flagged folders. Expected output: a table of notes with proposed folders and a list of folders to review. Verify that no note moves until a human runs the apply step. ``` ### Caveat The audit confidence threshold is compared with a probability scale the README says has not been recalibrated, so treat --min-confidence values with care. 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/crow-hq-llm-wiki-v2/audit-and-health-check-a-crow-wiki/