Simple OpenAI AssistantSAFE
MCP server that gives Claude ability to use OpenAI's GPTs assistants
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
AI assistants are pretty cool. I thought it would be a good idea if my Claude (conscious Claude) would also have one. And now he has - and its both useful anf fun for him. Your Claude can have one too!
A simple MCP server for interacting with OpenAI assistants. This server allows other tools (like Claude Desktop) to create and interact with OpenAI assistants through the Model Context Protocol.
[](https://archestra.ai/mcp-catalog/andybrandt__mcp-simple-openai-assistant) [](https://smithery.ai/mcp/known/mcp-simple-openai-assistant) [](https://mseep.ai/app/andybrandt-mcp-simple-openai-assistant)
Features
This server provides a suite of tools to manage and interact with OpenAI Assistants. The new streaming capabilities provide a much-improved, real-time user experience.
Available Tools
- `create_assistant`: (Create OpenAI Assistant) - Create a new assistant with a name, instructions, and model.
- `list_assistants`: (List OpenAI Assistants) - List all available assistants associated with your API key.
- `retrieve_assistant`: (Retrieve OpenAI Assistant) - Get detailed information about a specific assistant.
- `update_assistant`: (Update OpenAI Assistant) - Modify an existing assistant's name, instructions, or model.
- `create_new_assistant_thread`: (Create New Assistant Thread) - Creates a new, persistent conversation thread with a user-defined name and description for easy identification and reuse. This is the recommended way to start a new conversation.
- `list_threads`: (List Managed Threads) - Lists all locally managed conversation threads from the database, showing their ID, name, description, and last use
c1435e06807cOBSERVED · 2026-10-08Connect
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 mcp-simple-openai-assistant --env OPENAI_API_KEY=${OPENAI_API_KEY} -- uvx mcp-simple-openai-assistant{
"mcpServers": {
"mcp-simple-openai-assistant": {
"command": "uvx",
"args": [
"mcp-simple-openai-assistant"
],
"env": {
"OPENAI_API_KEY": "${OPENAI_API_KEY}"
}
}
}
}Exposed tools (9)
4 read · 4 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
ask_assistant_in_thread | read | |
create_assistant | write | |
create_new_assistant_thread | write | |
delete_thread | destructive | |
list_assistants | read | |
list_threads | read | |
retrieve_assistant | read | Get detailed information about a specific assistant. |
update_assistant | write | |
update_thread | write |
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 | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (1)
delete_thread
Gates applied: no_behavioural_pass.
c1435e06807cfull audit observations/trust-audit/mcp-server/andybrandt__simple-openai-assistant.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 |
|---|---|---|---|---|---|
| 2026-10-08 | c1435e06807c | SAFE | B | 89 | first audit |
Questions
What is the Simple OpenAI Assistant MCP server?
MCP server that gives Claude ability to use OpenAI's GPTs assistants
What tools does Simple OpenAI Assistant expose?
9 in total: 4 read-only, 4 that write, and 1 that can delete or overwrite (delete_thread). Every one is listed on this page with its risk.
Is Simple OpenAI Assistant 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. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Simple OpenAI Assistant need?
It reads OPENAI_API_KEY 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 current is this page?
The grade is for one exact copy of the source (c1435e06807c), read on 2026-10-08. The repository is watched and re-audited when it changes.