What it does
This MCP server acts as a runtime firewall for AI agents. According to its description, it combines pattern-based and heuristic detection to scan prompts, RAG documents and tool arguments. The README presents it as a runtime guard for item #1 of the OWASP LLM Top 10, which covers prompt injection.
It suits developers whose agents take in external text, such as retrieved documents or tool inputs, and who want that content checked before the agent acts on it.
The project is released under the MIT license and hosted on GitHub. The source gives no detailed setup steps; the README's quick-start section has no content.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CSOAI-ORG/agent-prompt-injection-firewall-mcp
The WAF for agents. Pattern-based + heuristic firewall scans prompts, RAG documents, tool argume...
v1.0.13As shown on registry.modelcontextprotocol.io when we last read it