Back to skill

Security audit

Vision Tool

Security checks for vulnerabilities and agentic risk

Overview

The skill is a straightforward image-recognition wrapper that sends a user-chosen image to a local Ollama service, with no hidden persistence or unrelated data access found.

Install only if you are comfortable with selected images and prompts being sent to your configured Ollama service. Keep Ollama local for sensitive images, and review development cleanup commands before running them in a modified workspace.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · CONTRIBUTING.md (reported line 93)May include surrounding context.

md
pip install ruff pytest pytest-cov

# Clean build files
rm -rf build/ dist/ *.egg-info/
find . -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true
find . -type f -name "*.pyc" -delete
rm -rf .pytest_cache/ htmlcov/ .ruff_cache/

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · CONTRIBUTING.md (reported line 96)May include surrounding context.

rm -rf build/ dist/ .egg-info/ find . -type d -name "pycache" -exec rm -rf {} + 2>/dev/null || true find . -type f -name ".pyc" -delete rm -rf .pytest_cache/ htmlcov/ .ruff_cache/

text

### Quick Reference

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill documentation advertises use of the Ollama /api/chat endpoint and therefore network-capable behavior, but it does not declare any explicit tool scope such as permissions or allowed-tools. Missing capability declarations weaken least-privilege controls and make it harder for users or the host platform to understand and constrain what the skill can do.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill processes user-supplied images by sending them to the Ollama /api/chat service, but the description does not clearly disclose that image content is transmitted for external or service-side analysis. This creates a privacy and consent risk because users may provide sensitive images without realizing they are being forwarded beyond the immediate tool invocation.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% 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 · scripts/vision_core.py (reported line 94)May include surrounding context.

python
start_time = time.time()

        try:
            response = requests.post(url, json=payload, timeout=120)
            elapsed = time.time() - start_time

            if response.status_code == 200:

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The entry "Default English prompt "Describe this image"" indicates the skill defaults to English, which can be a natural-language locale policy concern when no user opt-in or language-selection behavior is mentioned. In this markdown file, there is no accompanying note that users can choose another language or that the English default is required for a documented reason.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The feature list states "English prompts" and presents this as the expected mode of operation, which is a language-policy constraint in natural language. The file does not offer an opt-in choice for other languages or explain a justified region-specific requirement.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The text specifies a default English prompt of "Describe this image," which reinforces a fixed language choice without documenting user opt-in or necessity. This can be a natural-language policy issue when the skill appears to assume one language by default.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This code sends the user-provided prompt and base64-encoded image contents to the Ollama /api/chat endpoint via an HTTP request. Although the module docstring mentions Ollama, there is no explicit user-facing disclosure at the point of execution that the image data will be transmitted to a service endpoint.

Content

No source excerpt is available for this finding.