Back to skill

Security audit

Task Routing

Security checks for vulnerabilities and agentic risk

Overview

This is a local task-routing helper with no hidden network, credential, persistence, or command-execution behavior.

This skill appears safe to install for local task routing and prioritization. Review any custom routing rules you load, since rule data controls which TaskMetadata fields are compared and what target labels are returned.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (2)

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The code substantially matches the declared purpose of a task routing engine: it loads routing rules, evaluates task metadata, and returns a routing decision. It uses change type, size, and risk as described. However, the declared description explicitly includes urgency as a routing factor, and the provided code never evaluates any urgency field or condition. Additionally, the implementation is a deterministic rule-based router rather than a broader 'intelligent' assignment engine, though that alone is a weaker discrepancy. There are no suspicious undeclared permissions, external resource access, or unrelated triggers. The main mismatch is the missing urgency-based routing capability.

Content

No source excerpt is available for this finding.

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · src/models.py (reported line 99)May include surrounding context.

python
def matches(self, task_metadata: TaskMetadata) -> bool:
        """Check if condition matches task metadata."""
        task_value = getattr(task_metadata, self.field, None)

        if self.operator == "equals":
            return task_value == self.value