Versioned security assessment

Report ID: SA-EC4623A7

6/30/2026, 8:26:10 AM

azure-hosted-copilot-sdk security assessment v2

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
azure-hosted-copilot-sdk
Version
v2
Maintainer
microsoft
Source
github.com/microsoft/github-copilot-for-azure/tree/main/plugin/skills/azure-hosted-copilot-sdk/
Coverage
5 Files scanned · 403 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

Medium

3 confirmed security findings require attention.

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 analysis found many command, network, filesystem, and credential patterns, but review shows they are documentation for legitimate Azure and GitHub Copilot SDK deployment workflows. No prompt injection or confirmed malicious exfiltration was found. The skill remains medium risk because it guides agents through cloud deployment commands, token handling, template copying, and remote documentation lookups.

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

5 Files scanned · 403 Lines analyzed

3 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.

  1. Source

    Binding unavailable

  2. Artifact

    Identity incomplete

  3. Audit

    Complete

  4. 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.

Observed in 2 evidence locations

Network access

May connect to external services.

Observed in 5 evidence locations

Filesystem access

May read or write local files.

Observed in 3 evidence locations

Env variables

May read values from the process environment.

Observed in 6 evidence locations

External commands

May invoke commands or programs outside the Skill.

Observed in 9 evidence locations

Risk findings

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

Confirmed security concerns (3)

RISK-001 Medium
SKILL.md:20SKILL.md:46SKILL.md:51references/copilot-sdk.md:65-67references/deploy-existing.md:9-11references/azure-model-config.md:96-98
Cloud Deployment Commands Require User Review
The skill instructs agents to run Azure Developer CLI, Docker, GitHub CLI, Azure CLI, and curl commands. These are legitimate for building and deploying a Copilot SDK app, but they can change cloud resources, start local services, or send test requests if executed without user confirmation.
The command instructions are present and operationally meaningful. They are expected for Azure deployment, so the risk is legitimate but not malicious.
RISK-002 Medium
references/deploy-existing.md:37-44references/deploy-existing.md:59-61references/azure-model-config.md:61-69references/azure-model-config.md:76-82references/existing-project-integration.md:40-43
Credential and Token Handling Workflow
The skill documents GitHub token retrieval, Azure credential use, Key Vault storage, and Azure model endpoint variables. This is necessary for the deployment scenario, but mishandling could expose tokens or configure privileged identities incorrectly.
The files explicitly discuss tokens, managed identity, Key Vault, and credential environment variables. Context shows legitimate use, but the impact of mistakes is security relevant.
RISK-003 Medium
references/deploy-existing.md:3-5references/deploy-existing.md:15-25references/deploy-existing.md:27-35references/deploy-existing.md:46-48
Template Copying Can Modify User Projects
The skill tells agents to copy infrastructure, scripts, Dockerfiles, and azure.yaml from a scaffolded template into a user project. It warns against running scaffolding inside the user project, but the workflow still requires careful review to avoid overwrites or unintended deployment configuration.
The file modification workflow is documented directly. The skill includes a mitigation warning, reducing severity but not eliminating operational risk.

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
references/copilot-sdk.md:15-22references/azure-model-config.md:53-54references/azure-model-config.md:62references/existing-project-integration.md:65-67
Hardcoded Documentation URLs Are Benign
The hardcoded URLs point to GitHub Copilot SDK documentation, Azure endpoint examples, localhost tests, and Azure cognitive services scopes. I did not find evidence that these URLs exfiltrate credentials or target unrelated infrastructure.
The URL contexts are documentation links, endpoint placeholders, localhost testing, and an Azure token scope. No suspicious destination or secret-bearing request was found.
Low
SKILL.md:18-22references/azure-model-config.md:9-12references/copilot-sdk.md:5-7references/existing-project-integration.md:26-36
Static Backtick Findings Are Markdown False Positives
Most external command findings are caused by markdown inline code and fenced examples, not hidden executable scripts. The skill content is instructional documentation, not code that executes automatically.
The reviewed locations are markdown examples or identifiers. They do not execute unless an agent or user chooses to run the documented commands.

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