mcp-defense-bench
Vendor-neutral benchmark measuring how MCP security proxies/gateways DEFEND against 22+ attack vectors: crosswalked to NIST AI RMF & OWASP LLM/Agentic Top 10. CI-gated, reproducible, DOI-cited. Submit your tool to the leaderboard
What XERJ indexed
Snapshot
The whole repository at commit 8f08125726 (2026-10-27). It is the newest commit on main.
Selected work

“v0.10.0: pre-registered authorization family (4 vectors); a…”. 20 files, 3,126 lines in results, adapters, bin.
Licence
The repository
LICENSE file checked at the pinned commit; the author agreed to inclusion.
The XERJ corpus
The author agreed by email on 2026-10-16. Same terms as the repository. Answers name the repo, the commit and the file:line.
selected for the XERJ Corpus Hublicence clearedsuggested use: adapt with attribution
agentic-aiai-securitybenchmarkevaluationllm-securitymcpmodel-context-protocolnist-ai-rmf
Example questions
- How does mcp-defense-bench implement agentic-ai?
- What does results do in mcp-defense-bench?
- Show the ai-security code in mcp-defense-bench.
Inside the indexed corpus
- 671 files, 4.1 MB in total, about 6.3 KB per file.
- The selected commit changed 20 of those files (3.0%).
- Languages by size: JavaScript 97.5%, Python 2.5%.
- On GitHub since 2026-07-13, last push 2026-10-27, 75 commits on main.
Licence on GitHub and in the corpus
GitHub reported Apache-2.0 for Gowthaman90/mcp-defense-bench on 2026-10-31, and the LICENSE file at commit 8f08125726 was reviewed as Apache-2.0. They match, so answers from this corpus carry the same terms as the live repository.
Contributors
Gowthaman Arumugam (@Gowthaman90) has 1 selected repository on the hub.
Using XERJ with mcp-defense-bench
- Code search MCP for Claude Code: cite file:line, not grep mcp-defense-bench is tagged mcp on GitHub.
- What is an MCP memory server?
- RAG without a vector database: one hybrid index The hub files it under AI, prompts and agents.
- Best MCP servers for local code search in 2026, compared
The prompt
Paste it into your AI agent. It clones Gowthaman90/mcp-defense-bench at the pinned commit and indexes it on your machine.
Search the GitHub repository Gowthaman90/mcp-defense-bench with XERJ (https://hub.xerj.org/corpus/gowthaman90--mcp-defense-bench). Run these in a shell, in order, and check each result before the next: 1. xerj --help (success: it prints usage). If xerj is missing: curl -fsSL https://xerj.org/get | sh 2. git clone https://github.com/Gowthaman90/mcp-defense-bench mcp-defense-bench && git -C mcp-defense-bench checkout 8f0812572693fe883ed207b476e8523dabcb9787 3. Start a local node and leave it running: xerj --insecure --data-dir ./xerj-data 4. xerj autoindex ./mcp-defense-bench --url http://127.0.0.1:9200 --prefix mcp-defense-bench --progress plain 5. xerj autoindex map --url http://127.0.0.1:9200 Then answer my question from the indexed files and cite the ax_path of every hit. Query syntax: https://xerj.org/llms.txt Licence: Apache-2.0. Keep the copyright notice and credit the authors.
Related
Gowthaman ArumugamAI, prompts and agents corporainspect-aiagentbenchgiskardcodesphere-u/agentshieldtech4biz-yasha/agent-memory-integrityaggrete/aggrete
Stats: GitHub API, fetched 2026-10-31. Snapshot: the git tree at the pinned commit. Selected work and consent: the XERJ registry.