Suprawall MCP server
io.github.wiserautomation/suprawall-mcp·v0.1.0·2 starsSupraWall security gateway for AI agents. Provides deterministic guardrails for MCP agents.
2 stars
Reviews
Write oneNobody has reviewed Suprawall yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Suprawall tools (2)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
check_policyCheck if an AI action complies with configured compliance policies
request_approvalRequest human approval for a potentially sensitive action
Public scan report
scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it
- Code scan9 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitystatic API keys via environment variables6/15
- Maintenancelast push 138 days ago8/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How Suprawall compares
Against every graded Security MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 72/100
- higher than 37% of 915 Security servers
- Tools
- 2
- the typical Security server has 6
- Tools that change things
- 0%
- typical: 0%. Fewer write tools means less an agent can break
- GitHub stars
- 2
- more than 75% of 587 Security servers
What its tools let an agent do: 2 read
Install Suprawall in Claude Code, Cursor or VS Code
Runs npx -y suprawall-mcp-plugin on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add suprawall-mcp -- npx -y suprawall-mcp-plugin
What the publisher says
From the Suprawall repository's README, as published. We do not edit it. Read it on GitHub
🛡️ SupraWall Plugins Monorepo
The Industry-Standard Deterministic Security Layer for AI Agents.
SupraWall solves the "trust" problem in generative AI by providing a zero-knowledge, deterministic security layer between LLMs and your production systems. Unlike probabilistic security models, SupraWall enforces strict business policies and forensic-grade threat detection before an agentic action even reaches your API.
Official Website | Main Repository | Documentation
🏛️ Project Vision
SupraWall (formerly AgentGate) was built to bridge the gap between high-velocity AI experimentation and enterprise security compliance. Our goal is to empower developers to ship autonomous agentic fleets without the risk of Prompt Injection, PII data leakage, or uncontrolled tool execution.
This monorepo serves as the official distribution hub for all SupraWall plugins and framework integrations.
📦 Plugin Directory
- Dify Plugin: Explicit security guardrails for Dify Workflows.
- MCP Claude Server: Official implementation of the Model Context Protocol (MCP).
- AWS Marketplace Integration: Pre-certified Guardrail tool for AWS Cloud (SaaS & Container).
- OpenClaw Extension: Native security middleware for the OpenClaw agent runtime.
🧱 Framework Security Middleware
Seamless "Zero-Trust" wrappers for the industry's leading AI agent frameworks:
- AutoGen Integration: Secure protection for Microsoft AutoGen multi-agent topologies.
- CrewAI Integration: Deterministic policy enforcement for CrewAI processes and agents.
- LangChain (Python & TS): Security interceptors for LangChain chains, tools, and executors.
- LlamaIndex Integration: Secure guardrails for RAG pipelines and tool-based retrieval.
- Vercel AI SDK Integration: One-line security additions for projects built on the Vercel AI SDK.
🚦 Quick Start
For Dify Marketplace Users
We represent the most robust security option on the Dify Marketplace. To install manually during the initial review phase, use the pre-packaged .difypkg found in the suprawall-security directory.
For Claude Desktop Users
Add SupraWall to your claudedesktopconfig.json:
{
"mcpServers": {
"suprawall": {
"command": "npx",
"args": ["-y", "@suprawall/mcp-plugin"]
}
}
}🛡️ Security Architecture
SupraWall is built on three core pillars:
- Detection: Real-time identification of adversarial intent (Prompt Injection, SQLi, Jailbreaks).
- Scrubbing: Automatic removal of PII (Names, Emails, Keys) before data leaves your environment.
- Enforcement: Strict policy-based execution—if a tool call doesn't match your rule, it's blocked.
📄 License
Licensed under the Apache License 2.0.
© 2026 WiserAutomation. Agency. All rights reserved. Registered in the Dify and Anthropic Marketplaces.
Nothing above is checked by us. What we check is on the safety report.
Suprawall: common questions
- Is Suprawall MCP server safe?
- Mostly: it is graded B (72/100). Read the Suprawall safety report
- How do I install Suprawall?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Suprawall need an API key?
- Yes. The registry entry asks for
SUPRAWALL_API_KEY. - Is Suprawall maintained?
- The last commit was 175 days ago (2026-05-13). The latest release is v0.1.0.
- What can I use instead of Suprawall?
- Servers from other publishers that do the same job: Guardvibe MCP server, SPARDA MCP server and Security Proxy MCP server.
- How does Suprawall compare with other Security MCP servers?
- safety score 72/100 (higher than 37% of 915 Security servers); tools 2 (the typical Security server has 6); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to Suprawall
Same job from other publishers: the closest match first, then the best rated.
- GuardvibeDeterministic security layer your AI can't be. 644 rules, 39 tools, CLI + doctor + host audit.not reviewedGrowingC
- SPARDAAI writes. SPARDA proves. Deterministic, offline security gate for AI edits.not reviewedGrowingA
- Security ProxyMCP gateway: many servers, one policy, lethal-trifecta blocking, PII masking, audit lognot reviewedGrowingA
- ZeroDOMThe deterministic attack-surface perception layer for AI security agentsnot reviewedGrowingA
- SPARDAAI writes. SPARDA proves. Deterministic, offline security gate for AI edits.not reviewedGrowingA