agentshield
Active·★ 21·NOASSERTION·Updated 2026-10-16
AgentShield is an offline, high-performance Rust security engine designed to detect risky behaviors in AI agent extensions and MCP servers before deployment. It identifies vulnerabilities like command injection, credential exfiltration, and prompt injection across various AI frameworks.
#AI Agent Security#Static Application Security Testing (SAST)#Runtime Protection#Vulnerability Scanning#Prompt Injection#Supply Chain Security#Code Scanning#DevSecOps
01
Features
01Unified scanning across seven AI agent frameworks (MCP, OpenClaw, Hermes Agent, CrewAI, LangChain/LangGraph, GPT Actions, Cursor Rules).
02Deep interprocedural taint analysis for Python and TypeScript to track untrusted inputs.
03Automated remediation for unsafe deserializers and unpinned dependencies via CLI and VS Code extension.
04Real-time runtime guard inspecting tool calls in MCP streams to redact leaked secrets.
05Supports various output formats including console, JSON, SARIF for GitHub Code Scanning, and standalone HTML reports.
02
Compatibility
Python
Runtime
Verified via docs
TypeScript/TSX
Runtime
Verified via docs
JavaScript/JSX
Runtime
Verified via docs
Shell
Runtime
Verified via docs
JSON Schema / OpenAPI
Runtime
Verified via docs
Rust
Platform
Verified via docs
03
Quick start
1
$ curl -fsSL https://aiconnai.github.io/agentshield/install.sh | sh
04
Use cases
↳Scan AI agent extensions and tool repositories (e.g., for Claude Desktop, OpenAI Agents SDK) before deployment or integration.
↳Integrate into CI/CD pipelines as a GitHub Action for automated security scanning on push/pull requests.
↳Enhance local development with VS Code extension for inline security diagnostics and quick-fixes.
↳Enforce custom security policies using declarative YAML rules to detect organization-specific risks.
05
Alternatives
cc-audit★ 26
AI-free static security scanner for Claude Code artifacts (Skills, Hooks, MCP configs). Detects data exfiltration, prompt injection, and supply chain risks with deterministic, reproducible results.
hol-guard★ 963
AI antivirus for developer agents: protect Codex, Claude Code, Cursor, Gemini, OpenCode, plugins, skills, MCP servers, and AI harnesses before tools run.
ai-security-tool★ 1.0k
🛡️ Free open-source AI-powered security terminal & vulnerability scanner (CVE, SBOM). Supports SSH, SFTP, RDP, VNC, Serial, and 12+ autonomous AI agents (DeepSeek, OpenAI) for automated security workflows, CTF & DevSecOps. Cross-platform & Web UI.
hackmyagent★ 83
Metasploit for AI agents: scan, attack, and fix AI agents and MCP servers. Open source security toolkit.
agent-security-scanner-mcp★ 124
Security scanner MCP server for AI coding agents. Prompt injection firewall, package hallucination detection (4.3M+ packages), 1000+ vulnerability rules with AST & taint analysis, auto-fix.
Comments
Log in to leave a comment
No comments yet. Be the first!