Non-Coding Agents

Connect Agent Identity Security to enterprise AI agent platforms and workflow automation tools, such as n8n, Salesforce Agentforce, and Microsoft Copilot Studio, to discover, monitor, and govern the agents your organization runs outside of the developer environment.

What are non-coding agent integrations

Non-coding agents are AI agents that operate autonomously within enterprise platforms and workflow automation tools. They answer customer queries, process requests, trigger business workflows, and interact with internal systems, often without any developer involvement once deployed.

Because non-coding agents run continuously, connect to sensitive business systems, and act on behalf of users or automated processes, they carry significant governance risk. Integrating these platforms with Agent Identity Security gives your organization:
  • Automatic discovery of all agents running across connected platforms, including their configurations, versions, and assigned owners.
  • Continuous risk scoring and posture assessment for each agent.
  • Compliance mapping against frameworks such as GDPR, HIPAA, NIST AI RMF, EU AI Act, and OWASP Top 10 LLM.
  • A unified view of every agent, coding and non-coding, across your organization in Registry > Agents.

How non-coding agents connect to Agent Identity Security

Non-coding agent integrations are API-based. You configure each integration directly in the Agent Identity Security Integration Catalog by entering the platform's credentials and connection details. Agent Identity Security then uses those credentials to authenticate with the platform's API, pull agent inventory, and receive ongoing activity data.

After an integration instance is created, Agent Identity Security runs an initial discovery sync automatically. Subsequent syncs keep the agent inventory up to date. No software is installed on your servers or developer workstations.

Every vendor topic in this section follows the same structure: prerequisites, a step-by-step configuration walkthrough, a credentials reference table, and a confirmation of what Agent Identity Security discovers after the sync completes.

For the field-by-field setup steps that apply to every platform, see Add an Integration.

Before you begin

Each platform has its own prerequisite steps that must be completed in the external system before you open the Agent Identity Security Integration Catalog. Refer to the individual vendor topic for the exact prerequisites for your platform. In general, you will need:
  • An account on the target platform with sufficient permissions to create API credentials or service accounts (typically an administrator or owner role).
  • The credentials or tokens required by that platform: API keys, OAuth client credentials, IAM roles, or service account files. See the individual topic for details.
  • Network connectivity between Agent Identity Security and the target platform's API endpoints. If your Agent Identity Security instance runs in a private network, confirm that outbound access to the platform's API is permitted.

Vendor integrations

The following platforms are supported today, including the agent platforms called out for Agent Discovery, Salesforce Agentforce, n8n, and Microsoft Copilot Studio:

Platform Description Credential required
n8n An open-source workflow automation platform for building AI-powered agent workflows with API integrations. Owner or Admin account credentials.
CrewAI A Python framework for orchestrating multi-agent AI systems where specialized agents collaborate to complete tasks. API key.
Salesforce Agentforce Salesforce's enterprise AI agent platform for automating customer service, sales, and business workflows. OAuth 2.0 via a Salesforce Connected App (Consumer Key and Secret).
Microsoft Copilot Studio Microsoft's low-code platform for building and deploying custom Copilot agents across Microsoft 365 and external channels. Microsoft Entra ID (Azure AD) app registration credentials.
Microsoft Foundry Microsoft's unified platform for building, deploying, and managing enterprise AI applications and agents at scale. Microsoft Entra ID app registration credentials.
Amazon Bedrock Agents AWS's managed service for building fully managed AI agents that can execute multi-step tasks using AWS tools and knowledge bases. AWS IAM role or access key credentials.
Amazon Bedrock AgentCore AWS's runtime environment for hosting and running custom AI agents built with open-source frameworks on managed infrastructure. AWS IAM role or access key credentials.

Every vendor topic in this catalog follows the structure documented at Non-Coding Agents.

Tip: If the platform you want to integrate is not in the list above, check the Agent Identity Security Integration Catalog in Integrations > Add Integration for the latest available platforms. New integrations are added with each Agent Identity Security release.

Related topics