Versioned security assessment

Report ID: SA-9B24CAFD

6/30/2026, 7:50:22 AM

azure-upgrade security assessment v2

Skill Security Certification Report

Audit History

Audit model: codex Historical report

Skill name: azure-upgrade
Version: v2
Maintainer: microsoft
Source: github.com/microsoft/azure-skills/tree/main/.github/plugins/azure-skills/skills/azure-upgrade/
Coverage: 6 Files scanned · 955 Lines analyzed
Policy version: Unavailable

Confirmed finding summary

No confirmed security findings

The completed audit recorded no confirmed security findings. This is not proof that the Skill has no side effects.

Installation context

Historical evidence

This report may not describe the currently installable artifact. Open the current Skill page for install guidance.

Open current Skill page

This report does not block or authorize the manifest or ZIP.

Static command-execution findings are mostly true in context: this skill is documentation that instructs an AI agent to generate and run Azure CLI, curl, and deployment commands. I found no prompt-injection attempt or confirmed malicious exfiltration, but the skill can create, modify, deploy, download packages from, and delete Azure resources, so it should not be published without stronger execution safeguards.

Report position

Historical report

Open audit history before using this report to install.

Audit attestation

Not attestable

The required immutable binding is incomplete.

Human verification

Not verified

No human verification is recorded for this report.

Coverage

6 Files scanned · 955 Lines analyzed

4 items shown for review

Limitations

This report does not claim runtime or sandbox execution and does not prove the absence of side effects.

Evidence chain

Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.

Source

Binding unavailable
Artifact

Identity incomplete
Audit

Complete
Install contract

Open manifest to verify
Open manifest

Capabilities observed

Observed means this report recorded supporting evidence. Not recorded does not prove that a capability is absent.

Contains scripts

May execute code included with the Skill.

Not recorded by this audit

Network access

May connect to external services.

Observed in 4 evidence locations

Filesystem access

May read or write local files.

Observed in 4 evidence locations

Env variables

May read values from the process environment.

Not recorded by this audit

External commands

May invoke commands or programs outside the Skill.

Observed in 6 evidence locations

Capability review items (4)

High

references/services/functions/automation.md:231-238 references/services/functions/automation.md:322-332 references/services/functions/automation.md:402-407 references/services/functions/automation.md:415-420

Agent-Guided Azure Resource Modification and Deletion

The skill guides execution of Azure CLI commands that create a new function app, deploy code, restart apps, and delete function apps. The documents include confirmation language, but marketplace use would still expose users to high-impact cloud changes if an agent executes the steps with broad Azure permissions.

The referenced lines contain concrete Azure CLI commands for migration, deployment, restart, and deletion. The intent is legitimate Azure upgrade automation, but the operational impact is clearly high.

High

references/services/functions/automation.md:109-111 references/services/functions/automation.md:151-171 references/services/functions/automation.md:212-222

Sensitive Configuration and Package Retrieval

The automation guide collects app settings, reads AzureWebJobsStorage, lists managed identity role assignments, and downloads a deployment package using a storage connection string. These steps can expose secrets or proprietary application code in the workspace or command output.

The commands directly retrieve app settings, identity data, and deployment packages. This is expected for migration work, but it creates real exposure risk if logs or files are mishandled.

Medium

SKILL.md:57 references/services/functions/automation.md:352-358 references/services/functions/automation.md:371-375 references/services/functions/consumption-to-flex.md:139-141

Network Calls to User Workload Endpoints

The skill instructs the agent to call the default Azure Functions hostname and HTTP trigger endpoints with curl. This is a normal validation step, but it can contact production endpoints and may trigger application behavior.

The curl commands and validation step are explicit. The risk depends on the target app behavior, so this is moderate rather than inherently malicious.

Medium

SKILL.md:61 references/workflow-details.md:17-19 references/services/functions/assessment.md:22-28 references/services/functions/automation.md:220-222

Local Workspace File Creation for Upgrade Artifacts

The skill requires creating status and assessment files in the workspace and can download a deployment package. This is useful for migration tracking, but it may persist sensitive configuration or application artifacts locally.

The files and downloaded package are specifically named. The sensitivity depends on what values the agent writes into reports or package files.

Risk findings

Confirmed security concerns are separated from items that still need review.

No confirmed security findings were recorded for this completed audit.

Expert evidence

Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.

Artifact subject

Marketplace commit: Unavailable
Content hash: Unavailable
Tree hash: Unavailable
Skill path: Unavailable
Audit payload hash: Unavailable

Analysis metadata

Audit model: codex

Analysis state: Complete

Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.

Static false positives ignored (2)

Low

SKILL.md:12 references/global-rules.md:26-39 references/workflow-details.md:9-13 references/services/functions/assessment.md:30-117

Static Weak-Crypto and Backtick Hits Are Markdown False Positives

Many weak cryptography and Ruby backtick findings map to Markdown emphasis, diagrams, inline code, Azure SKU names, or shell examples. I found no executable Ruby code, hidden cryptographic implementation, or prompt-injection instruction in the reviewed files.

The lines are documentation constructs or code-fenced templates, not executable skill code. The broader command examples remain risky, but these specific pattern classes are false positives.

Low

references/services/functions/automation.md:6 references/services/functions/automation.md:317-318 references/services/functions/consumption-to-flex.md:223-227

Hardcoded Documentation URLs Are Benign References

The hardcoded URLs point to Microsoft Learn and Azure sample documentation. I did not find evidence that the skill sends user data or secrets to those URLs.

The URLs are cited as documentation links, not request targets for collected data. Network activity in the skill comes from Azure CLI and curl validation commands instead.

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest

The manifest and lockfile bind install artifacts to cryptographic hashes. This integrity claim is separate from the security assessment.

Audit attestation: not_attestable