CrewAI Integrations

AI Agents Directory is the world's largest AI agents marketplace and directory, offering over 1,300 enterprise-ready AI agents across 64+ categories. It provides a comprehensive platform to discover, compare, and deploy AI agents tailored to various business needs. Users can explore a wide range of categories, including productivity, sales, customer service, coding, voice, and more, each populated with specialized agents designed to automate tasks and enhance efficiency. It features detailed information on each agent, allowing for informed decision-making based on features, pricing, and reviews. With tools to request custom AI agents and submit new ones, AI Agents Directory fosters a dynamic ecosystem for businesses and developers seeking to leverage AI solutions.

Naptha is a modular AI platform for autonomous agents that empowers developers and researchers to build, deploy, and scale cooperative multi‑agent systems on the agentic web. Its core innovations include Agent Diversity, which continuously upgrades performance by orchestrating diverse models, tools, and architectures; Horizontal Scaling, which supports collaborative networks of millions of AI agents; Self‑Evolved AI, where agents learn and optimize themselves beyond human‑designed capabilities; and AI Agent Economies, which enable autonomous agents to generate useful goods and services. Naptha integrates seamlessly with popular frameworks and infrastructure, LangChain, AgentOps, CrewAI, IPFS, NVIDIA stacks, and more, via a Python SDK that upgrades existing agent frameworks with next‑generation enhancements. Developers can extend or publish reusable components on the Naptha Hub, run full agent stacks anywhere a container can execute on Naptha Nodes.

Atla is the agent observability and evaluation platform that dives deeper to help you find and fix AI agent failures. It provides real‑time visibility into every thought, tool call, and interaction so you can trace each agent run, understand step‑level errors, and identify root causes of failures. Atla automatically surfaces recurring issues across thousands of traces, stops you from manually combing through logs, and delivers specific, actionable suggestions for improvement based on detected error patterns. You can experiment with models and prompts side by side to compare performance, implement recommended fixes, and measure how changes affect completion rates. Individual traces are summarized into clean, readable narratives for granular inspection, while aggregated patterns give you clarity on systemic problems rather than isolated bugs. Designed to integrate with tools you already use, OpenAI, LangChain, Autogen AI, Pydantic AI, and more.

Gentoro is a platform built to empower enterprises to adopt agentic automation by bridging AI agents with real-world systems securely and at scale. It uses the Model Context Protocol (MCP) as its foundation, allowing developers to automatically convert OpenAPI specs or backend endpoints into production-ready MCP Tools, without writing custom integration code. Gentoro takes care of runtime concerns like logging, retries, monitoring, and cost optimization, while enforcing secure access, auditability, and governance policies (e.g., OAuth support, policy enforcement) whether deployed in a private cloud or on-premises. It is model- and framework-agnostic, meaning it supports integration with various LLMs and agent architectures. Gentoro helps avoid vendor lock-in and simplifies tool orchestration in enterprise environments by managing tool generation, runtime, security, and maintenance in one stack.

Silmaril is a self-healing prompt injection defense designed to protect AI systems from increasingly complex, multi-step attacks that traditional guardrails fail to stop. It operates by wrapping inference calls and evaluating whether an execution sequence is leading toward a harmful outcome, rather than simply filtering inputs. It uses a multihead classifier that analyzes user intent, application context, and execution states together, enabling it to detect indirect injection, multi-turn attack chains, context poisoning, and tool abuse before damage occurs. Silmaril continuously strengthens its defenses through autonomous threat hunting agents that probe systems, discover vulnerabilities, and generate synthetic training data from real attack scenarios. These insights are used to retrain the model automatically, deploying updated protections in under an hour and propagating anonymized defenses across all deployments.

Singulr is an enterprise AI governance and security platform that provides a unified control plane to help organizations discover, secure, and optimize AI adoption at scale. It addresses the growing gap between rapid AI usage and limited governance by delivering complete visibility into all AI systems in use, including homegrown applications, embedded AI, public tools, and shadow AI that often remains invisible to security teams. It continuously discovers and inventories AI assets across the organization, creating a real-time map of agents, models, and services, while assessing their risk through contextual analysis of data handling, model lineage, vulnerabilities, and compliance implications. Through its Singulr Pulse intelligence layer, it evaluates millions of AI systems, assigns risk scores, and supports automated onboarding workflows that reduce approval cycles from weeks to hours without compromising security.

GuardionAI is an Agent and MCP Security Gateway that provides unified security for AI agents and Model Context Protocol tools operating on enterprise data. It sits in the execution path to discover, redact sensitive data, enforce protection, and give teams visibility into actions that traditional SIEM, DLP, and identity layers cannot see. Every agent action is inspected, enforced, and logged at the protocol level across AI agents, LLM apps, RAG systems, chatbots, coding agents, MCP servers, internal tools, databases, operating systems, and cloud environments. GuardionAI protects against critical AI threats such as prompt injection, system override, web attacks, MCP tool poisoning, malicious code execution, NSFW content, PII and credential exposure, confidential data leakage, off-topic drift, and unauthorized access, mapped to OWASP LLM Top 10 and agentic AI threat frameworks. Its gateway provides four layers of protection.

Pillar Security is a unified AI security platform for securing the agentic workforce across the entire AI lifecycle, from development to deployment and runtime protection. It connects business context across discovery, testing, and protection so security intelligence compounds across AI applications, agents, models, prompts, frameworks, tools, MCP servers, skills, coding agents, SaaS, cloud, code, and endpoints. Pillar helps organizations discover and manage AI assets everywhere, including shadow AI and unapproved systems, assess supply chain and posture risks, map agentic attack surfaces, and validate the vulnerabilities that actually matter. Its AI Security Posture Management capabilities analyze connected agents, tools, permissions, data sources, prompts, models, and supply chain components to expose risky paths, policy violations, misconfigurations, coding agent risks, and blast radius when a single component is compromised.

Vulnify is runtime authorization for AI agents. Before a sensitive tool call runs, your code calls check() and gets ALLOW, REVIEW or BLOCK, with a 0-100 risk score and the reasons: action, data sensitivity, destination, volume. REVIEW pauses only that action until a person approves. Start in monitor mode, fail closed by default. Policies as code, Node and Python SDKs (MIT) with LangChain, MCP, CrewAI and LangGraph adapters, and a hash-chained audit log with JSON/CSV export. Free plan available.