agentic-guard is an open-source static analysis tool that scans LLM agent code for prompt injection and confused-deputy vulnerabilities. Below are 13 code review & quality apps with similar functionality to agentic-guard, matched by what each product actually does — not ranked or scored. Explore each to find the closest fit for your use case.
ai-agentguard is an open-source CLI tool that monitors AI coding agents for security threats such as remote code execution, MCP poisoning, and API key theft. It is designed for developers working with AI agent frameworks to enhance security.
AI-native quality gate for agent-generated code — scan, audit, auto-fix, trend
AgentGuard is a free and open-source local desktop application designed to provide guardrails for AI coding agents. The tool focuses on enhancing security and efficiency by controlling and monitoring prompts and requests made by AI agents. Its primary features are Guardian Agent, which offers data loss prevention on prompts, tool inputs, and file reads, and Token Saver, which compresses verbose outputs and reduces unnecessary token usage before requests are sent to the AI model. Guardian Agent intercepts requests to block secrets, risky dependency installs, and oversized requests, applying token caps and flagging vulnerabilities in dependencies. It also provides update advice for outdated dependency pins. Token Saver reduces the context size by compressing shell outputs, caching repeated file reads, and shrinking noisy responses from search, diff, or JSON operations. These optimizations are tracked, with logs and analytics available to review token savings and event histories. AgentGuard includes review surfaces such as Logs, Analytics, and Garden. Logs provide a compact, exportable event history detailing Guardian and Token Saver actions. Analytics offer charts that track models used, token totals, savings, and trends in latency and tool calls. The Garden feature visualizes project activity, mapping modules, files, symbols, and import relationships. The application supports integration with three AI coding agents: Claude Code, Codex CLI, and Cursor. Guardian Agent's controls extend across all three integrations, while Token Saver offers the most extensive support for Claude Code. Delivery options include downloadable versions for Mac and Ubuntu, with the option to build from source using npm and Tauri for other operating systems. AgentGuard is positioned as a tool for developers seeking local data loss prevention, dependency protection, token optimization, and observability for workflows involving AI coding agents.
atomicguard is an open-source agent framework designed to improve the reliability of LLM-based code generation. It uses guard-validated loops to ensure generated code meets safety and correctness criteria, making it suitable for developers building autonomous coding agents or tools that require robust code validation.
AgentGuard is an MCP server designed to provide quality assurance for AI-generated code, particularly from agents such as Claude, GPT, or Cursor. It enables a structured five-step process that guides AI agents to produce production-ready code, focusing on ensuring consistency, type safety, and robust validation. The tool addresses common issues in AI-generated code, such as inconsistent file structures, incomplete error handling, partial typing, low test coverage, disorganized imports, and lengthy manual refactoring. A central feature of AgentGuard is its use of archetypes, which are architectural templates tailored to specific project needs. These archetypes help standardize project structure, enforce type-safe interfaces and schemas, and automate the implementation of validation logic. The tool also performs structural and security checks and generates quality reports with trace summaries, making every validation decision explainable and traceable. AgentGuard offers a marketplace with over 61 archetypes and includes 8 built-in archetypes and 19 MCP tools, supporting a range of project types and validation requirements. AgentGuard is delivered as a Python package, installable via pip, and does not require API keys or extra configuration. The tool is licensed under the MIT License and is available for free, as indicated by the "Get Started Free" prompt. AgentGuard is developed by RLabs SpA. This platform is intended for developers and teams seeking to automate and enhance the quality assurance process for AI-generated code. By providing real-time validation, granular verification mechanisms, and comprehensive quality reports, AgentGuard aims to reduce manual QA time and improve the reliability of code produced by AI agents.
agentguardCI is an open-source command-line tool that provides contract testing for tool-using AI agents. It integrates with CI pipelines to automatically evaluate and validate agent behaviors, helping developers ensure reliability and correctness of their AI systems.
agentlint is an open-source CLI tool that provides real-time quality guardrails and linting for AI coding agents. It helps developers maintain code quality and safety by integrating with agentic coding workflows and enforcing best practices.
AgentGuard is a runtime governance platform designed for production AI agents, with a focus on meeting regulatory and compliance requirements in financial services, particularly for APRA-regulated organizations in Australia and the Asia-Pacific region. The tool provides operational controls and audit evidence aligned with frameworks such as APRA CPS 230, the EU AI Act, and ISO 42001, enabling organizations to deploy AI agents that their regulators can approve. A central feature of AgentGuard is its runtime enforcement: it operates within the agent execution environment, intercepting tool calls, model outputs, and spending actions before they occur. Policy enforcement is managed through YAML-based declarative policies, allowing organizations to define tool whitelists, daily spending limits, requirements for human approval, and operational rules such as after-hours restrictions. The platform generates signed, Bitcoin-anchored audit records for every agent decision, and compiles these into compliance packs in PDF format, mapped to regulatory frameworks and suitable for board and auditor review. AgentGuard integrates with multiple AI agent frameworks, including OpenClaw, LangChain, CrewAI, AutoGen, OpenAI Assistants, and MCP. js with a single decorator or middleware. For OpenClaw, AgentGuard provides native governance features such as per-agent policy, fleet-level monitoring, cost tracking, and a killswitch for policy breaches, all without requiring code forks or patches. The platform can be installed locally or behind an organization's firewall, with no dependency on a SaaS runtime. The service is available under several pricing tiers: a free plan for one agent with limited retention and an MIT-licensed SDK; a Team plan at $499 AUD per month for up to 25 agents and extended retention with additional support; and a Compliance plan at $2,500 AUD per month, offering unlimited agents, extended retention, compliance evidence packs, and Bitcoin-anchored audit roots. An enterprise option with on-premises deployment and dedicated support is also available. AgentGuard is developed by The Bot Club Pty Ltd.
agenticlint is an open-source command-line linter that analyzes OpenAPI specifications to ensure they are safe and clear for use as MCP or AI-agent tools. It performs static analysis to identify potential issues, helping developers maintain secure and well-documented APIs.
agent-memory-guard is an open-source CLI tool that provides runtime defense for AI agent memory, protecting against poisoning, tool abuse, and privilege escalation. It is designed for AI security engineers and follows OWASP guidelines.
inter-agent-guard is an open-source security firewall library designed for multi-agent AI systems. It helps developers protect their agent-based applications from prompt injection and other security threats, with features like LangGraph integration and multi-agent support.
aport-agent-guardrails is an open-source core library that provides pre-action authorization and security guardrails for AI agent and LLM frameworks. It helps developers enforce policies and secure agent actions before execution.
agnostic-security is an open-source CLI and API tool that acts as a firewall for AI coding agents, preventing the leakage of secrets, PII, and credentials. It integrates with popular AI coding tools and supports DevSecOps workflows.
