Practical overview
When it fits
Use it when you want to run retrieval over code and documentation on infrastructure you manage, with access through a browser, HTTP API or MCP connector.
Who it is for
- Developers who want to search a codebase together with its documentation.
- Teams building a self-hosted knowledge collection from files such as PDFs, Markdown and text.
- MCP-aware agent users who want to connect an agent to the knowledge system.
Documented uses
- Ingest documents through uploads or scans of a mounted folder.
- Ask semantic questions across ingested material and inspect citations.
- Run keyword searches without generating an LLM answer.
- Expose retrieval to an MCP-aware agent through the included connector.
Requirements & limits
The documented quick start runs the stack locally with Docker Compose and opens the interface at localhost:8080.
- The documented stack uses ChromaDB, Docling, FastAPI and Docker Compose.
- The default model and embedding provider is Ollama; the README also lists OpenAI, Anthropic, Gemini and OpenAI-compatible provider options.
- Folder ingestion relies on a host directory mounted through
DOCS_DIR; uploads also support PDF, Markdown and text files. - The MCP connector is a companion component, separate from the core web UI and API.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up the self-hosted stack
Start the Docker Compose services and open the local web interface.
Before you start
- Docker Compose
- A local copy of the project repository
Steps
Clone the repository and change into its directory.
Copy
.env.exampleto.envand adjustDOCS_DIRand LLM provider settings as needed.Run the command below.
docker compose up -dCheck
http://localhost:8080/health, then openhttp://localhost:8080.
Expected outcome
The project instructions direct users to the local UI and provide a health endpoint for checking the backend.
Guide limitations
- The guide documents startup and access, not a completed ingestion or query workflow.
- The service uses the configured LLM and embedding providers.
Compare options for Retrieval-augmented applications
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Knowledge Base Self Hosting Kit | Ingest documents through uploads or scans of a mounted folder. | Docker, Self-hosted | The documented quick start runs the stack locally with Docker Compose and opens the interface at localhost:8080. |
| Rag Powered AI Assistant Transforming Data Retrieval And Analysis Across The Web And Pdfs | Retrieve relevant passages from web and PDF content to answer a question. | Web | |
| LangChain in Supabase Edge Functions | Initialize and start a Supabase project locally. | Cloud, Server |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 16 October 2026
- Original website title
- 2dogsandanerd/Knowledge-Base-Self-Hosting-Kit · Official homepage title · retained in its source language

