100Critical
This skill executes powerful cloud commands, but its reliance
Behavior Analysis
Description verified — behavior matches claim
Claims to do
Azure Deploy: > **AUTHORITATIVE GUIDANCE — MANDATORY COMPLIANCE** > > **PREREQUISITE**: The **azure-validate** skill **MUST** be invoked and completed with status `Validated` BEFORE executing this skill.
Actually does
This skill reads a deployment plan from `.azure/deployment-plan.md` and executes Azure deployments. It runs `azd up`, `azd deploy`, `terraform apply`, and `az deployment` commands, utilizing MCP tools like `mcp_azure_mcp_azd` and `azure__role` to interact with Azure. It performs pre-deployment checks, RBAC health checks, and post-deployment verification, including querying Azure for provisioned roles and presenting `https://` endpoint URLs.
Skill Info
Namemicrosoft/github-copilot-for-azure/azure-deploy
Registrygithub
Version4081320c2592
PURLpkg:github/microsoft/github-copilot-for-azure@4081320c2592?skill=azure-deploy
Stars202
Installs154,300
Source
SHA-256174e366b4a2d...
SHA-1aaa38651755b...
MD53a4cf7422924...
TLSH31d1629a7b56...
ssdeep192:pfuQK+0U...
Size6,633 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
GitHub
https://github.com/microsoft/github-copilot-for-azure/tree/main/skills/azure-deploy154,300 installs
Skills.shDocs
npx skills add https://github.com/microsoft/github-copilot-for-azureAssessments (2)
AI Agent Traps ↗Broad Command Execution Capabilitiescriticalcommand execution
The skill is explicitly designed to execute powerful Azure CLI, Terraform, and AZD commands, including `azd up`, `azd deploy`, `terraform apply`, and `az deployment`. These commands can make significant, destructive, or costly changes to cloud infrastructure.
100% confidenceline 3
This skill runs azd up, azd deploy, terraform apply, and az deployment commands with built-in error recovery. ... MCP Tools: mcp_azure_mcp_azd
Supply Chain Risk from External Markdown Referencesmediumsupply chain
The skill relies on several external markdown files (e.g., `Pre-Deploy Checklist`, `global-rules.md`, `recipes/README.md`) for instructions and rules. If these external references are not securely managed or can be modified by an attacker, they could be used to inject malicious instructions or bypass safety mechanisms.
90% confidence
Pre-deploy checklist required — [Pre-Deploy Checklist](references/pre-deploy-checklist.md) ... Load Recipe — Based on recipe.type in .azure/deployment-plan.md
Potential for Bypassing User Confirmationmediumautonomy abuse
While the skill states 'Destructive actions require `ask_user`', it also includes a step to run `azd provision --no-prompt`. This flag bypasses user interaction, which could be abused if the provision step leads to unintended resource creation or modification without explicit human approval.
80% confidence
⛔ Destructive actions require ask_user ... RBAC Health Check — For Container Apps + ACR with managed identity: run azd provision --no-prompt
Extensive Reconnaissance Capabilitieslowreconnaissance
The skill has access to tools for listing Azure subscriptions, resource groups, and role assignments (`mcp_azure_mcp_subscription_list`, `mcp_azure_mcp_group_list`, `azure__role`). While intended for verification, these tools can be used to enumerate cloud resources for malicious purposes.
100% confidence
MCP Tools: mcp_azure_mcp_subscription_list | mcp_azure_mcp_group_list | azure__role
Reliance on Mutable Configuration for Executionhighmemory poisoning
The skill's execution flow, including which commands to run and their parameters ('Load Recipe'), is heavily dependent on the content of `.azure/deployment-plan.md`. If this file is compromised or manipulated, it can lead to arbitrary command execution or unintended infrastructure changes by bypassing validation checks.
100% confidence
Read .azure/deployment-plan.md, verify status = Validated ... Load Recipe — Based on recipe.type in .azure/deployment-plan.md
Version History (2)
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/microsoft/github-copilot-for-azure/azure-deploy)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/microsoft/github-copilot-for-azure/azure-deploy"><img src="https://mondoo.com/ai-agent-security/api/badge/github/microsoft/github-copilot-for-azure/azure-deploy.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/microsoft/github-copilot-for-azure/azure-deploy.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow