Back to skill

Security audit

SrbijaVoz Timetable and Announcements

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently checks public Srbija Voz transit information and does not show deceptive, destructive, credential-seeking, or persistence behavior.

Before installing, understand that the skill runs a bundled Python script that contacts public Srbija Voz websites and saves lookup results to a local JSON file unless configured otherwise. It should be used for public transit-status checks, not for account access or protected timetable systems.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill instructs the agent to run a Python script that performs network access, but it does not declare any explicit tool scope such as allowed-tools or permissions. That creates an authorization gap where the agent may execute networked code with broader capabilities than intended, reducing enforceability and auditability of what the skill is permitted to do.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 52)May include surrounding context.

python3 scripts/srbvoz_scraper.py --station "Beograd"

text

Return the best matching station names and codes. Do not guess station names without checking the API first.

## Inspect Timetable Metadata

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The file is entirely structured around Serbian-language terms for notice classification and does not indicate any user choice, opt-in, or locale scoping. This can be a language/locale policy concern if the broader skill is expected to operate across languages but silently assumes Serbian input.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.