GPT-5SAFE
GPT-5 Local MCP
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
A Model Context Protocol (MCP) server that provides seamless integration with OpenAI's GPT-5 API through Claude Code. This server enables you to leverage GPT-5's advanced capabilities directly within your Claude Code workflows.
🚀 Features
- Direct GPT-5 Integration: Call GPT-5 API with simple prompts or structured conversations
- Two Powerful Tools:
gpt5_generate: Simple text generation with promptsgpt5_messages: Structured conversation handling with message arrays- Built for Claude Code: Optimized for seamless integration with Anthropic's Claude Code IDE
- TypeScript Support: Fully typed for better development experience
- Error Handling: Robust error handling with detailed feedback
- Usage Tracking: Built-in token usage reporting
📋 Prerequisites
- Node.js (v18 or higher)
- OpenAI API key with GPT-5 access
- Claude Code IDE
🛠 Installation
1. Clone the Repository
git clone https://github.com/AllAboutAI-YT/gpt5mcp.git cd gpt5mcp
2. Install Dependencies
cd servers/gpt5-server npm install
3. Build the Server
npm run build
4. Configure Environment Variables
Create a .env file in the servers directory:
# servers/.env OPENAI_API_KEY=your-openai-api-key-here
🔧 Claude Code Integration
Add the Server to Claude Code
claude mcp add gpt5-server -e OPENAI_API_KEY=your-openai-api-key-here -- node /path/to/gpt5mcp/servers/gpt5-server/build/index.js
Verify Installation
Test the server with a simple query:
Ask GPT-5: "Hello, how are you today?"
📚 Available Tools
gpt5_generate
Generate text using a simple input prompt.
Parameters:
input(required): The text prompt for GPT-5model(optional): GPT-5 model variant (default: "gpt-5")instructions(optional): System instructions for the modelreasoning_effort(optional): Reasoning level ("low", "medium", "high")- `max_toke
4aa5e92d10d2OBSERVED · 2026-10-07Connect
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 gpt5-server --env OPENAI_API_KEY=${OPENAI_API_KEY} -- npx -y [email protected]{
"mcpServers": {
"gpt5-server": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"OPENAI_API_KEY": "${OPENAI_API_KEY}"
}
}
}
}Exposed tools (2)
2 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
gpt5_generate | read | Generate text using OpenAI GPT-5 API with a simple input prompt |
gpt5_messages | read | Generate text using GPT-5 with structured conversation messages |
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 | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (2 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (3)
const envPath = path.join(__dirname, '../../.env');
dotenv, node-fetch, zod, zod-to-json-schema, @types/node, typescript
Gates applied: no_behavioural_pass, no_license.
4aa5e92d10d2full audit observations/trust-audit/mcp-server/allaboutai-yt__gpt-5-1.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 |
|---|---|---|---|---|---|
| 2026-10-07 | 4aa5e92d10d2 | SAFE | B | 89 | first audit |
Questions
What is the GPT-5 MCP server?
GPT-5 Local MCP
What tools does GPT-5 expose?
2 in total: 2 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is GPT-5 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 GPT-5 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 does GPT-5 run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as gpt5-server at 0.1.0.
How current is this page?
The grade is for one exact copy of the source (4aa5e92d10d2), read on 2026-10-07. The repository is watched and re-audited when it changes.