Works with the AI Agents you already use
azure activity logs threat analysis
Analyze Azure logs, build KQL hunting queries, and generate SOC-ready threat reports for cloud environments.
$15
azure activity logs threat analysis
Example session with this skill installed
Analyze these AzureActivity log exports from the last 24 hours. I suspect unauthorized privilege escalation in the production subscription. Provide a hunting report and KQL for validation.
- Read your context and instructions
- Compiled the azure activity logs
I have analyzed the provided AzureActivity logs for the production subscription over the last 24 hours. The analysis focused on administrative role assignments and high-impact control plane changes.
Azure Activity Log Threat Analysis Report
| Finding | Evidence / Operation | Actor | Impact | Confidence |
|---|---|---|---|---|
| Suspicious Role Assignment | Microsoft.Authorization/roleAssignments/write | svc_deploy@contoso.com | Owner permissions granted to a new service principal. | High |
| MFA Requirement Bypass | Update user (conditional access modification) | admin_hunter@contoso.com | Reduced security posture for privileged accounts. | Medium |
| Key Vault Access Change | Microsoft.KeyVault/vaults/accessPolicies/write | svc_deploy@contoso.com | Potential secret exfiltration capability. | High |
Validation KQL Queries
// Search for new Owner or Contributor assignments in the last 24 hours
AzureActivity
| where TimeGenerated > ago(24h)
| where OperationNameValue == "Microsoft.Authorization/roleAssignments/write"
| extend Properties_d = parse_json(Properties)
| extend RoleDefinition = tostring(Properties_d.roleDefinitionId)
| where RoleDefinition contains "8e3af657-a8ff-443c-a75c-2fe8c4bcb635" // Owner Role
| project TimeGenerated, Caller, RoleDefinition, ResourceGroup, _ResourceId
Detection Gaps
- Gap: No active alert for 'Owner' role assignments outside of known PIM windows.
- Gap: Lack of correlation between Service Principal creation and immediate permission assignment.
Next steps
- Verify if
svc_deploy@contoso.comwas part of a scheduled deployment. - Review the audit logs for
admin_hunter@contoso.comto check for session hijacking signs. - Implement the provided KQL as a scheduled analytic rule in Microsoft Sentinel.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Security teams often struggle to parse through massive volumes of Azure Activity and Entra ID logs during incidents. Manually writing complex KQL queries while under pressure leads to missed indicators of compromise or inaccurate threat assessments.
What it does
- Generates KQL threat hunting queries for Azure control plane, identity, and Key Vault activity.
- Analyzes provided log exports to identify suspicious administrative changes and privilege escalation attempts.
- Produces structured threat hunting reports and incident support summaries based on evidence.
- Maps findings to MITRE ATT&CK techniques and identifies detection gaps in current monitoring.
- Converts log analysis into tool-agnostic detection logic or platform-specific SIEM content.
Frameworks & tools
Azure Activity Logs, Microsoft Entra ID (Azure AD), Kusto Query Language (KQL), Azure Log Analytics, and Azure Monitor.
Why this beats prompting it yourself
Standard LLM prompts often hallucinate KQL schemas or suggest offensive techniques. This skill enforces defensive boundaries, ensures schema accuracy for specific tables like SigninLogs or AzureDiagnostics, and follows a structured SOC-ready reporting methodology that separates facts from hypotheses.
Use cases
- Investigating a suspected credential theft via Entra ID sign-in anomaly logs.
- Building a detection gap register for a new Azure subscription deployment.
- Creating a timeline of administrative operations during a cloud security audit.
- Developing custom KQL alerts for unauthorized Network Security Group changes.
Known limitations
Requires authorized log data or specific workspace context to be useful. Does not perform live tenant interrogation without pre-configured access and credentials.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Requires manual provision of log data or workspace access. Will not provide offensive attack execution guidance. SIEM-specific output requires the user to name the target platform.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 months ago
- Software
Python
Permissions
No special permissions declared or detected