Skip to content

Add MCPGoat — a deliberately vulnerable Model Context Protocol (MCP) target - #276

Open
SabyasachiDhal wants to merge 1 commit into
OWASP:masterfrom
SabyasachiDhal:add-mcpgoat
Open

Add MCPGoat — a deliberately vulnerable Model Context Protocol (MCP) target#276
SabyasachiDhal wants to merge 1 commit into
OWASP:masterfrom
SabyasachiDhal:add-mcpgoat

Conversation

@SabyasachiDhal

@SabyasachiDhal SabyasachiDhal commented Aug 17, 2026

Copy link
Copy Markdown

Adds MCPGoat to the directory — a deliberately vulnerable Model Context Protocol (MCP) server for hands-on penetration-testing practice.

As AI agents adopt MCP, MCP servers are a fresh, under-tested attack surface. MCPGoat implements 26 challenges spanning MCP-specific attacks (tool poisoning, tool shadowing, rug-pulls, indirect prompt injection, sampling abuse) and classic web bugs (RCE, SSRF, SQLi, path traversal), each at Easy / Moderate / Difficult plus a Secure reference level — 78 capture-the-flag flags — with a victim-agent harness that shows a real LLM agent being exploited.

Conformance to CONTRIBUTING.md / schema.json:

  • Single entry added to _data/collection.json, tab-indented.
  • Validated against schema.json (all keys, enums, and required fields).
  • Sorted case-insensitively by name (lands between "Marathon" and "Metasploitable 2").
  • It is a vulnerable target, not a tools-only entry. Categorised ctf + single-player; collection = offline + container (ships a Dockerfile / docker compose).

🤖 Generated with Claude Code

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
@psiinon

psiinon commented Aug 17, 2026

Copy link
Copy Markdown
Member

You clearly have not searched for MCP on https://vwad.owasp.org/ 😁
Right now there are 4 others...

Comment thread _data/collection.json
Comment on lines +2433 to +2435
"TypeScript",
"Node.js",
"Docker"

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seem like this should include: MCP, LLM, AI

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Still outstanding

@SabyasachiDhal SabyasachiDhal changed the title Add MCPGoat — first Model Context Protocol (MCP) target in the directory Add MCPGoat — a deliberately vulnerable Model Context Protocol (MCP) target Aug 22, 2026
@SabyasachiDhal

Copy link
Copy Markdown
Author

Ha, you're right — I searched the repo's collection.json rather than the live site and over-claimed. Apologies, I've dropped the "first" framing from the title and description.

MCPGoat's angle is depth: 26 challenges, each hardened across Easy / Moderate / Difficult plus a Secure reference level (78 flags), covering MCP-specific attacks (tool poisoning, rug-pulls, sampling abuse) alongside the classic web bugs, with a victim-agent harness that shows a real LLM agent being exploited. Happy to tweak the wording or category so it sits well alongside the others — thanks for taking a look!

@kingthorin

Copy link
Copy Markdown
Collaborator

If you've got a logo you should include it as well

@kingthorin

Copy link
Copy Markdown
Collaborator

Ugh, just noticed you're on the old repo.

Please join us at our new home: https://vwad.owasp.org

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants