All skills
lukemurraynz avatar

/azure-sre-agent

@2cc2455

Design, configure, review, and operate production-grade Azure SRE Agent capabilities: response plans, scheduled tasks, HTTP triggers, custom agents, autonomous and review workflows, approval guardrails, AMBA observability, source RCA, connectors, MCP, governance hooks, WAF reviews, AI Foundry posture, Digital Native governance, postmortem generation, and KT discipline.

Use this Skill: https://skilld.dev/gh/lukemurraynz/hve-agent-skills/azure-sre-agent

This session only. Nothing lands on disk.

referencessecurity-identity-rbac.md

≈859 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Security, Identity, and RBAC Reference

Production Azure SRE Agent designs must make identity and permission boundaries explicit.

Design Rules

  1. Prefer managed identity and secretless access.
  2. Use read-only scopes for diagnostics.
  3. Use narrowly scoped custom roles for remediation candidates.
  4. Separate connector permissions from Azure RBAC review.
  5. Treat OBO/user-delegated fallback as elevated risk that requires explicit human approval.
  6. Validate private-network and DNS paths before routing private workload incidents.
  7. Keep write-capable actions behind hooks, Review mode, and DA/PPA.
  8. Treat managed connector credentials as delegated credentials from the configuring user. Any agent user with access to the connector can invoke enabled operations through that credential.
  9. For private-network workloads, validate required domains, WebSocket access, DNS, proxy rules, and VNet-integrated execution before routing incidents through the agent.

Network Allow-List (verified 2026-08-10)

The agent's execution environment and portal require these domains for HTTP and WebSocket traffic (https://learn.microsoft.com/azure/sre-agent/network-requirements):

Domain Purpose
*.azuresre.ai Agent portal, API, real-time chat (WebSocket)
sre.azure.com Agent management portal
portal.azure.com Azure portal (Monitor, Logs, managed identity)
api.applicationinsights.io Application Insights query API
api.loganalytics.io / api.loganalytics.azure.com Log Analytics query API
*.ods.opinsights.azure.com Log Analytics ingestion
management.azure.com Azure Resource Manager
login.microsoftonline.com / *.login.microsoft.com Microsoft Entra ID auth

Zscaler and some corporate proxies block *.azuresre.ai by default: a blocked domain presents as "portal won't load / chat unresponsive". Add it to the firewall allow list before debugging anything else. The agent resource also exposes read-only outboundIpAddresses for egress allow-listing.

No private-endpoint support exists on Microsoft.App/agents (verified via the resource schema 2026-08-10; re-checked 2026-08-25); private connectivity is via VNet integration (preview), configured with the agent's vnetConfiguration.subnetResourceId property. Do not design around a private endpoint on the agent itself.

CMK: treat as unsupported until listed. Microsoft.App/agents does not appear in the official "Services that support customer-managed keys" matrix matrix (https://learn.microsoft.com/azure/security/fundamentals/encryption-customer-managed-keys-support, page updated 2026-08-03, checked 2026-08-25). An exhaustive official support matrix omitting the service is affirmative absence, not an open question; re-check only when the service appears there.

Review Questions

  • Which identity executes Azure actions?
  • Which identity accesses repos, incident platforms, and observability systems?
  • What happens if managed identity lacks permission?
  • Which actions can modify production state?
  • Where are connector tokens stored and rotated?
  • How are private endpoints reached and logged?
  • Which managed connector operations are set to Allow versus Ask, and can any of them run in Autonomous mode?
  • Can the agent portal, data-plane endpoint, Application Insights, and required Azure domains be reached through corporate proxies and firewalls?

Source: SKILL.md on GitHub

No alerts8d3 checks · Risk SAFE
  • Gen Agent Trust Hub8d

    The Azure SRE Agent skill provides a production-grade framework for managing Azure infrastructure using AI agents. It incorporates extensive safety documentation, approval-based hooks, and least-privilege role templates. The 'low' verdict is assigned due to the inherent risk of indirect prompt injection when the agent processes external incident data and source code, a necessary function for its SRE capabilities.

  • Socket8d

    No alerts

  • Snyk8d

    Risk: LOW · No issues

Signed by skilld at 2cc2455. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated last month
compatibility
Azure SRE Agent; GitHub Copilot agent skills; new projects only
Other metadata
metadata
{
  "last_verified": "2026-08-25",
  "version": "2.23.3",
  "risk": "critical",
  "last_updated": "2026-08-25"
}

README badge

README badge for lukemurraynz/hve-agent-skills/azure-sre-agent