Back to skill

Security audit

Windows Ai

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for running a local Windows AI cluster, but it asks users to install unpinned code and open an unrestricted inbound firewall port for an apparently unauthenticated local AI service.

Install only in a dedicated environment, verify the `ollama-herd` package/source before installing, and avoid the provided broad firewall command as written. If remote nodes are needed, restrict access to trusted private IPs or a private network profile and confirm the service has appropriate authentication before exposing port 11435.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (2)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:25
Finding

Unpinned Third-Party Package Installation

Content
View full analysis
Remediation
View remediation
`. - Use a lock file or requirements file with cryptographic hashes and install with hash enforcement. - Document the expected package index and prevent fallback to untrusted indexes. - Prefer a dedicated virtual environment rather than installing into a global Python environment. - Verify the package publisher, source repository, release signature, and build provenance before recommending installation. - Review package updates before changing the pinned version. - Instruct users not to perform installation from an elevated shell unless elevation is demonstrably required. ]]>

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:35
Finding

Unrestricted Inbound Firewall Rule for the AI Service

Content
View full analysis
Remediation
View remediation
``` - Warn users not to enable the rule on Public network profiles and to verify the service's listening interfaces before exposing the port. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 68)May include surrounding context.

print(response.choices[0].message.content)

text

### curl (PowerShell)

```powershell
# Windows AI chat

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 122)May include surrounding context.

Image generation

powershell
curl http://localhost:11435/api/generate-image `
  -d '{"model": "z-image-turbo", "prompt": "futuristic Windows desktop", "width": 1024, "height": 1024}'