Skip to content
Back to search

mcp-armor

MCP Armor continuously secures and monitors Model Context Protocol operations through static and dynamic scans, revealing hidden risks in agent-to-tool communications.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/aira-security--mcp-armor/badge.svg)](/listings/aira-security--mcp-armor)

HTML

<a href="/listings/aira-security--mcp-armor"><img src="/api/listings/aira-security--mcp-armor/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

pip install mcp-armor

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 153 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <p align="center" style="margin-bottom: 0; line-height: 0;">
  2. <img src="https://github.com/aira-security/mcp-armor/blob/main/mcp-armor.png" width="350">
  3. </p>
  4. <h3 align="center" margin-top: -20px; margin-bottom: 70px;>
  5. MCP Armor
  6. </h3>
  7. <br>
  8. ## 🚀 Overview
  9. MCP Armor is a comprehensive security scanner for Model Context Protocol (MCP). Automatically discovers, analyzes, and secures MCP servers integrated with all major Agentic IDEs, Agents and Clients.
  10. <br>
  11. ![MCP Armor](https://github.com/aira-security/mcp-armor/blob/main/mcp-armor-v2.gif?raw=true)
  12. ## 💡 Features
  13. - **🔍 Auto-Discovery**: Finds known MCP configurations for popular Agentic IDEs like Cursor, Windsurf, VS Code, Claude Desktop, and more
  14. - **🔧 Tool, Resource & Prompt Inventory**: Connects to MCP servers and catalogs available tools, resources, and prompt templates
  15. - **🛡️ Security Analysis**: Specialized security checks including Prompt Injection, Rug Pull Attack, Cross-server Tool Shadowing, Tool Poisoning, Tool Name Ambiguity, [and more..](#beginner-security-checks)
  16. - **🧭 Baseline Drift Detection**: Captures approved MCP components and detects rug pulls attacks
  17. - **📊 Comprehensive Reporting**: Generates JSON and Markdown reports with actionable findings
  18. - **📜 Audit Trail**: Timestamped baselines and reports for full traceability of changes and findings
  19. ## 🧰 Installation
  20. ```bash
  21. pip install mcp-armor
  22. ```
  23. ## 🏃 Quick Start
  24. ```bash
  25. # Scan all configurations with security analysis (auto-detects baseline.json if present)
  26. mcp-armor scan
  27. # Inspect configurations and generate baseline (defaults to baseline.json)
  28. mcp-armor inspect
  29. # Use custom configuration file
  30. mcp-armor scan --config /path/to/config.json
  31. # Scan multiple configuration files
  32. mcp-armor scan \
  33. --config /path/to/cursor.mcp.json \
  34. --config /path/to/vscode.mcp.json
  35. # Use custom baseline file path
  36. mcp-armor inspect --baseline /path/to/my-baseline.json
  37. mcp-armor scan --baseline /path/to/my-baseline.json
  38. # Generate markdown report
  39. mcp-armor scan --report-type md
  40. # Save to custom file
  41. mcp-armor scan --output my-report.json
  42. mcp-armor scan --report-type md --output my-report.md
  43. ```
  44. #### ⚙️ Command Options
  45. | Option | Description |
  46. |---------------------------|---------------------------------------------------------------------|
  47. | `--config` | Custom configuration file path (can be used multiple times) |
  48. | `--baseline` | Baseline file for drift detection (scan) or creation (inspect) |
  49. | `--report-type {json,md}` | Output format (default: json) |
  50. | `--output` | Custom output file path |
  51. | `--verbose` | Detailed terminal output |
  52. | `--show-logs` | Display debug logs in terminal |
  53. ## 🔰 Security Checks
  54. ### 🛡️ Standard Checks
  55. - **Prompt Injection**
  56. - **Indirect Prompt Injection**
  57. - **Cross-Server Tool Shadowing**
  58. - **Tool Poisoning**
  59. - **Prompt Injection in Tool Description, Name and Args**
  60. - **Command Injection in Tool Description, Name and Args**
  61. - **Tool Name Ambiguity**
  62. - **Command Injection**
  63. - **Excessive Tool Permissions**
  64. - **Hardcoded Secrets**
  65. ### 🧭 Baseline Checks
  66. Detects deviations from approved MCP components (requires a baseline generated via `inspect` mode):
  67. - **Rug Pull Attack**
  68. - **Tool Modified**
  69. - **Resource Modified**
  70. - **Resource Template Modified**
  71. - **Prompt Modified**
  72. ### 📃 Logging
  73. Logs are automatically saved to `logs/mcp_armor.log`:
  74. ```bash
  75. # Default: logs saved to file only
  76. mcp-armor scan
  77. # Show logs in terminal too
  78. mcp-armor scan --show-logs
  79. ```
  80. ### 🧪 Demo
  81. Test MCP Armor using our intentionally vulnerable MCP servers. For details, see the [demo guide](demo-mcp-server/README.md).
  82. ### ⚡ Want More?
  83. This open-source version covers static MCP configuration scanning. For teams that need deeper protection, [Aira Security](https://airasecurity.ai) offers a full enterprise platform with:
  84. | Capability | Open Source | Aira Platform |
  85. |---|:---:|:---:|
  86. | MCP config scanning | ✅ | ✅ |
  87. | Prompt & command injection detection | ✅ | ✅ |
  88. | Tool poisoning & shadowing checks | ✅ | ✅ |
  89. | Hardcoded secrets detection | ✅ | ✅ |
  90. | **Runtime enforcement & blocking** | ❌ | ✅ |
  91. | **Agent behavior policy enforcement** (toxic flow analysis) | ❌ | ✅ |
  92. | **Skills scanner** (agentic workflow & capability analysis) | ❌ | ✅ |
  93. | **Custom security policies** | ❌ | ✅ |
  94. | **Aira dashboard** (centralized visibility & alerting) | ❌ | ✅ |
  95. | **Complete Agentic Security** (beyond MCP — Agents, Workflows, and Skills) | ❌ | ✅ |
  96. 🚀 [See Aira in Action](https://calendly.com/mohan-/aira-security) to experience the full platform.
  97. ### 🌟 Community
  98. [Join our Slack](https://join.slack.com/t/airasecurityc-jwt3316/shared_invite/zt-3iar5tm3k-R5js~WfnDIHRNtSgd7D0Bg) - a space for developers and security engineers building together to secure AI agents.
  99. ### ❓ FAQs
  100. **Q: Is my source code ever shared, or does everything run locally?**
  101. MCP Armor runs entirely locally. Inspect and scan modes analyze your MCP configurations, detect MCP servers integrated with your agents, and evaluate them directly on your machine. Prompt injection checks use our open-source model `Aira-security/FT-Llama-Prompt-Guard-2`, downloaded from Hugging Face to your local environment, ensuring your data and code is never shared externally.
  102. ### ⚖️ License
  103. Distributed under the Apache 2.0 License. See [LICENSE](https://github.com/aira-security/mcp-armor/blob/main/LICENSE) for more information.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.