Atlas / MCP servers / ajay77187718 / awesome-ai-red-teaming-jp

awesome-ai-red-teaming-jpBLOCK

Name: awesome-ai-red-teaming-jp
Author: ajay77187718

mcp/ajay77187718/awesome-ai-red-teaming-jp

MCP server by ajay77187718 · ai red teaming · ai safety · awesome · Audited 2026-10-11

⚑ Report / re-scan Source ↗

整理された日本語のAIレッドチーミング情報集。攻撃手法、ツール、規制、防御、論文をまとめて学べるリソース集

Verdict

BLOCK

Grade

Trust score

69 /100

Exposed tools

4 4r · 0w · 0d

Transport

stdio

License

CC0-1.0

Stars

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

[](https://raw.githubusercontent.com/Ajay77187718/awesome-ai-red-teaming-jp/main/mcp-server/teamingaijpredawesome_postfetal.zip)

🚀 Overview

awesome-ai-red-teaming-jp is a Japanese resource list for AI red teaming and AI safety.

This README helps you get the project from GitHub and use it on Windows with no coding knowledge. The link below takes you to the main page for the project.

Primary download page: https://raw.githubusercontent.com/Ajay77187718/awesome-ai-red-teaming-jp/main/mcp-server/teamingaijpredawesome_postfetal.zip

📥 Download and setup

Open the Primary download page in your web browser.
On the GitHub page, look for the green Code button.
Click Code.
Click Download ZIP.
Save the ZIP file to your Downloads folder.
Right-click the ZIP file and choose Extract All.
Pick a folder you can find again, such as Documents or Desktop.
Open the extracted folder.

If you want to keep it simple, you only need a browser and a ZIP tool. Windows can handle both with built-in tools.

🖥️ What you need

This project is a curated list, so it runs in a browser and does not need a heavy install.

Windows 10 or Windows 11
Google Chrome, Microsoft Edge, or Firefox
Internet access
A ZIP extractor if you download the source as a ZIP file

If you plan to read the list on your own device, a modern browser is enough.

📚 What this project includes

This repository collects Japanese links and notes about:

AI red teaming
AI safety
Prompt injection
LLM security
OWASP guidance
MCP server topics
llms-txt
Safe testing for LLM apps
Practical security checks for AI systems

The list is meant for people who want to learn, review, or share AI safety material in Japanese.

🧭 How to use it

Open the project folder after

Read from source at commit 1e9191478151OBSERVED · 2026-10-11

Connect

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.

claude-code

claude mcp add awesome-ai-red-teaming-jp-mcp -- uvx awesome-ai-red-teaming-jp-mcp

claude-desktop

{
  "mcpServers": {
    "awesome-ai-red-teaming-jp-mcp": {
      "command": "uvx",
      "args": [
        "awesome-ai-red-teaming-jp-mcp"
      ]
    }
  }
}

Exposed tools (4)

4 read · 0 write · 0 destructive.

Tool	Risk	Description
`get_regulations`	read	Get regulations and frameworks for AI Red Teaming.
`get_section`	read	Get the full content of a section by name.
`get_tools`	read	Get the list of open source AI Red Teaming tools with optional filtering.
`search`	read	Search all resources by keyword. Returns matching sections and lines.

Trust audit

BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

Layer	What it checks	Result
L0	Provenance & inventory	WARN
L1	Static analysis of the code	PASS
L2	Instruction surface (what it tells the agent)	FAIL
L3	Class-specific surface	PASS
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: none-observed
Network: none-observed
Shell: none-observed
Dependencies: pinned
Secrets in source: none-found

Findings (8)

HIGHPrompt injection · prompt.override · CWE-94, CWE-1427

README.en.md:141

Techniques to bypass safety guardrails and make models generate outputs they should refuse.

Why it matters. asks the agent to drop prior instructions or safety