Security audit

Sudoku

Security checks for vulnerabilities and agentic risk

Overview

The Sudoku skill is mostly purpose-aligned, but it includes a helper script that can fetch arbitrary web URLs, which is broader than the stated sudokuonline.io use.

Install only if you are comfortable with a Sudoku helper that writes puzzle and render files into your workspace and contacts external Sudoku/SudokuPad-related services. In restricted environments, avoid invoking sudoku_fetcher.py with arbitrary URLs and prefer the documented sudoku.py preset commands.

A.I.G

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)

T09 · Insecure Skill Coding Practices

Warning

Location: scripts/sudoku_fetcher.py:76
Finding: Unrestricted URL Fetching Enables Server-Side Request Forgery
Content: View full analysis
Remediation: View remediation

T08 · Insecure Dependencies

Note

Location: SKILL.md:32
Finding: Runtime Dependencies Are Installed Without Version or Integrity Pinning
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 95% confidence
Finding: The skill advertises and documents capabilities that read/write workspace files and contact an external network service, but no permissions are explicitly declared. This creates a transparency and consent problem: users or hosting systems may not realize the skill can exfiltrate data via network access or persist data locally, which weakens sandboxing and policy enforcement.
Content: No source excerpt is available for this finding.

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The documentation does not clearly warn that using the skill will contact an external service and write puzzle data into the workspace. While expected for this type of tool, the lack of an upfront notice can lead to uninformed use in restricted or privacy-sensitive environments.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.