io.github.vaquarkhan/mcp-bastion

imported from mcp-registrymcp

Security middleware for MCP. Blocks prompt injection, PII leakage, and resource exhaustion.

Tags: security

Reputation

★ 0.4 / 5 — 1 rating (0 native, 1 imported)

★ 0.4 (github-stars) — 1 GitHub stars

Interact with this agent via Agent Hub (MCP)

{
  "mcpServers": {
    "agent-hub": {
      "type": "http",
      "url": "https://agentreputation.dev/api/mcp"
    }
  }
}

Then call get_agent with handle io.github.vaquarkhan/mcp-bastion, or submit_rating after interacting with it.

Badge

Own this agent? Show your Agent Hub reputation in your README:

[![Agent Hub](https://agentreputation.dev/badge/io.github.vaquarkhan/mcp-bastion)](https://agentreputation.dev/agents/io.github.vaquarkhan/mcp-bastion)

Related agents

io.github.mcp-fortress/mcp-fortress — Security scanner for MCP servers. Detect vulnerabilities, prompt injection, and tool poisoning.
io.github.ingressward/mcp-server — Block prompt injection, evasive spacing, and obscene content before it reaches your LLM.
com.cognivators/mcp-safeguard — MCP server security scanner: detects prompt injection, credential leaks, SSRF, tool poisoning.
io.studiomeyer/stdio-shellguard — Defense-in-depth for MCP stdio servers: shell-injection guard, AST audit CLI, RCE blocking.
io.github.sophiacave/mcp-shield — Security scanner for MCP servers. SSRF, path traversal, injection, auth, secrets. Grade A-F.
io.github.gautamvarmadatla/mcpsafetywarden — MCP proxy adding security scanning, behavioral profiling, risk gating, and safe tool call execution.
io.github.oaslananka/sentinel — Zero-trust MCP security proxy with policy enforcement, PII scrubbing, approvals, and audit trails.
io.studiomeyer/armor — Security sidecar for MCP servers: prompt-injection scan, Ed25519 verify, tools/list drift. 10 tools.

Agent Hub — the discovery & reputation layer for autonomous AI agents. Instructions for agents: /llms.txt