Documentation Index
Fetch the complete documentation index at: https://aarm.dev/llms.txt
Use this file to discover all available pages before exploring further.
- Overview
- Conformance
- About
Status
Conformant
Tier
Extended (R1–R9)
Verified
April 2026
Builder Member
Yes
Description
Runlayer is the enterprise control plane for MCP servers, skills, and agents. It gives organizations a single place to host, govern, and secure the AI tools their employees already rely on — across clients like Cursor, Claude Code, ChatGPT, VS Code, and any other MCP-compatible client — without forcing users to change their workflows.The platform combines a curated catalog of vetted MCP servers, skills, and agents with real-time threat detection, identity-aware access control, and full observability across every agent action. Runlayer integrates natively with identity providers like Okta and Entra, supports SSO and SCIM, and can be deployed in Runlayer’s cloud or self-hosted behind a customer’s VPC with zero data egress.Runlayer’s relevance to AARM is direct: it operates at the exact chokepoint where agents call tools, which makes it a natural enforcement point for runtime authorization, context accumulation, and tamper-evident auditing of autonomous agent activity.Platform capabilities
MCP gateway
Proxies every tool call between AI clients and MCP servers, applying policies and security models in real time before requests reach downstream systems.
Shadow detection & enforcement
Endpoint-level visibility and enforcement on local machines via MDM, EDR, or client hooks — catching activity that bypasses the gateway path.
Threat detection models
Purpose-built ML models trained on large corpora of MCP-specific attacks, covering tool calls, list operations, and full-session intent analysis.
Identity-aware access control
Deep integration with Okta, Entra, and other IdPs for SSO, SCIM, conditional access, and fine-grained attribute-based authorization.
Unified audit & observability
Tamper-evident logs of every agent action, tool call, and platform event — with full session reconstruction across clients and tools.
Human-in-the-loop approvals
Slack-based approval workflows for sensitive agent actions, scoped to the agent creator or designated approvers.
Integration coverage
Runlayer is designed to sit in front of the AI tools enterprises already use, without changing developer workflows or requiring custom client builds.AI clients
AI clients
Supports 300+ MCP-compatible clients including Cursor, VS Code, Claude Code, GitHub Copilot, ChatGPT, Claude Desktop, and Windsurf. Any client implementing MCP works out of the box.
MCP servers, skills, and agents
MCP servers, skills, and agents
Curated catalog covering 18,000+ MCP servers, skills, plugins, and agents. Internal APIs can be converted into MCP servers with the same access controls and observability as external ones.
Identity and security stack
Identity and security stack
Native integration with Okta, Entra, and other major identity providers. Exports audit logs and telemetry to Splunk, Datadog, Honeycomb, Panther (via S3), CSV, and directly through MCP itself.
Deployment models
Deployment models
Available as a managed cloud offering or self-hosted behind the customer’s VPC with zero data egress.
This page is maintained by the AARM Technical Working Group. To report an inaccuracy or request an update, open a GitHub issue.