> For the complete documentation index, see [llms.txt](https://ai-security-docs.akto.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai-security-docs.akto.io/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents.md).

# AI Discovery Connectors

Connect Atlas to every surface where your employees run AI: CLIs, IDEs, browsers, SaaS AI suites, and the EDR/MDM tooling you already deploy through. Every connector feeds the same [Agentic AI Discovery](/akto-atlas-agentic-ai-security-for-employee-endpoints/ai-agent-activity.md) inventory: it doesn't matter which mix you pick, you get one view of everything discovered.

## Start With AI Endpoint Shield

For most fleets, [AI Endpoint Shield](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/ai-endpoint-shield.md) is the connector to deploy: it discovers local MCP servers, installs your CLI and IDE hooks, and bundles the System Proxy for standalone desktop apps, all through one fleet-wide rollout. Everything below either plugs into it or covers a surface it doesn't reach.

## Prefer a Lighter Touch?

If you're not ready for a fleet-wide rollout, or just want to cover one specific tool on a endpoint, set these up directly:

**Agentic CLI hooks**, zero-install, per tool:

* [Claude CLI Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/claude-cli-hooks.md)
* [Codex CLI Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/codex-cli-hooks.md)
* [Gemini CLI Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/gemini-cli-hooks.md)
* [Kiro CLI Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/kiro-cli-hooks.md)
* [Copilot Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/copilot-cli-hooks.md) - covers GitHub Copilot in VS Code and the CLI
* [Snowflake Cortex Code CLI Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/snowflake-cortex-cli-hooks.md)
* [Hermes Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/hermes-hooks.md)
* [Amp Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/amp-hooks.md)
* [OpenCode Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/opencode-hooks.md)

**Agentic IDEs and desktop apps:**

* [Cursor Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/cursor-hooks.md)
* [Neovim Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/neovim-hooks.md)
* [Claude Cowork Connector](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/claude-cowork-connector.md)

**Browser-based AI usage**, a surface AI Endpoint Shield doesn't cover at all:

* [Browser Extensions](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/browser-extensions.md) - discovery and inline guardrails for AI usage inside Chrome, Firefox, and Safari.

## Already Running an EDR?

Detect AI coding tools already running on managed endpoints through the EDR you deploy, no separate agent to install:

* [Deploy via SentinelOne](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/deploy-via-sentinelone.md)
* [Deploy via CrowdStrike](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/deploy-via-crowdstrike.md)
* [Deploy via Microsoft Defender Endpoint](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/deploy-via-microsoft-defender.md)

For MDM-based rollout instead (Intune, NinjaOne, Automox, Jamf, Mosyle), see the deployment options on the [AI Endpoint Shield](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/ai-endpoint-shield.md) page.

## SaaS and Compliance Connectors

These pull activity data straight from the vendor's own API, no device agent required:

* [Anthropic Connector](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/anthropic-connector.md) - pulls compliance activity from Claude.ai, Claude Console, and the Claude API via Anthropic's Compliance API.
* [Claude Inference Hooks](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/claude-inference-hooks.md) - use Atlas as the security server behind Anthropic's Inference Hooks, to allow or deny Claude prompts inline before inference runs.
* [OpenAI Connector](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/openai-connector.md) - pulls audit log and org activity data from ChatGPT and OpenAI via the OpenAI Admin API.
* [GitHub Copilot Enterprise](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/github-copilot-enterprise.md) - agent discovery, centralized guardrails, and enterprise-wide policy enforcement for Copilot.
* [Copilot Studio](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/microsoft-copilot-studio.md), [Copilot Studio (Multi Environment)](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/microsoft-copilot-studio-multi-environment.md), and [Copilot Studio (Threat Detection)](/akto-atlas-agentic-ai-security-for-employee-endpoints/endpoints-discovery-agents/microsoft-copilot-studio-threat-detection.md) - secure agents built in Microsoft Copilot Studio, across single or multiple Power Platform environments.

## Locally Hosted Models

Atlas also discovers models served by local model runners on employee devices, including Ollama, vLLM, SGLang, and Docker Model Runner.