MinimaSAFE
On-premises conversational RAG with configurable containers
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Minima is an open source RAG on-premises containers, with ability to integrate with ChatGPT and MCP. Minima can also be used as a fully local RAG or with your own deployed LLM.
Minima currently supports four modes:
- Isolated installation (Ollama) – Operate fully on-premises with containers, free from external dependencies such as ChatGPT or Claude. All neural networks (LLM, reranker, embedding) run on your cloud or PC, ensuring your data remains secure.
- Custom LLM (OpenAI-compatible API) – Use your own deployed LLM with OpenAI-compatible API (vLLM, Ollama server, TGI, etc.). The indexer runs locally while the LLM can be on your server, cloud, or local machine. No Ollama deployment needed, lighter resource usage, and full control over your LLM infrastructure.
- Custom GPT – Query your local documents using ChatGPT app or web with custom GPTs. The indexer runs on your cloud or local PC, while the primary LLM remains ChatGPT.
- Anthropic Claude – Use Anthropic Claude app to query your local documents. The indexer operates on your local PC, while Anthropic Claude serves as the primary LLM.
Running as Containers
Quick Start with run.sh
The easiest way to start Minima is using the run.sh script:
./run.sh
You'll see the following options:
Select an option: 1) Fully Local Setup (Ollama) 2) Custom LLM (OpenAI-compatible API) 3) ChatGPT Integration 4) MCP usage 5) Quit
Manual Docker Compose Commands
- Create a .env file in the project's root directory (where you'll find .env.sample). Place .env in the same folder and copy all environment variables from .env.sample to .env.
- Ensure your .env file includes the following variables:
LOCALFILESPATH EMBEDDINGMODELID EMBEDDINGSIZE OLLAMAMODEL (only for Ollama m
48445ee4a555OBSERVED · 2026-09-26Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add minima-app --env LLM_API_KEY=${LLM_API_KEY} --env PASSWORD=${PASSWORD} -- npx -y [email protected]{
"mcpServers": {
"minima-app": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"LLM_API_KEY": "${LLM_API_KEY}",
"PASSWORD": "${PASSWORD}"
}
}
}
}Exposed tools (1)
1 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
minima-query | read | Find a context in local files (PDF, CSV, DOCX, MD, TXT) |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (5 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (8)
favicon.icns
.env.sample
.gitmodules
@emotion/react, @emotion/styled, @mui/material, @mui/system, @testing-library/jest-dom, @testing-library/react, @testing-library/user-event, antd
electron
langchain, langchain-core, langfuse, langchain_qdrant, langchain_community, langchain-huggingface, langchain-text-splitters, transformers
httpx, google-cloud-firestore, firebase_admin, uvicorn
requests, ollama, langgraph, langchain, langchain-core, langchain_qdrant, langchain-ollama, langchain-openai
Gates applied: no_behavioural_pass.
48445ee4a555full audit observations/trust-audit/mcp-server/dmayboroda__minima.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-09-26 | 48445ee4a555 | SAFE | B | 89 | first audit |
Questions
What is the Minima MCP server?
On-premises conversational RAG with configurable containers
What tools does Minima expose?
1 in total: 1 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Minima safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Minima need?
It reads LLM_API_KEY and PASSWORD from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Minima run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as minima-app at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (48445ee4a555), read on 2026-09-26. The repository is watched and re-audited when it changes.