Back to skill

Security audit

ad-spend-performance-report

Security checks for vulnerabilities and agentic risk

Overview

The bundled skill has limited local access, but its report generator does not perform the advertised calculations.

Correct and independently validate CSV parsing, calculations, and attribution details before using reports for finance or budget decisions. Choose a fresh output directory to avoid overwriting an existing report. The separate Windows application is unnecessary for the Python workflow and is outside this inspection.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The code is related to advertising reporting, but its implemented behavior falls materially short of the declared purpose. The output reproduces supplied values rather than computing the promised performance metrics. Its generic guardrails do not provide the promised per-number attribution caveats. No unrelated access, external communication, or other undeclared capability is evident.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_write.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The generated report echoes input values and formula text instead of calculating ROAS and CPA, and does not attach attribution caveats to reported numbers.

Medium
Category
Excessive Agency
Confidence
100% confidence
Finding

The script presents a performance report but only echoes supplied values and formula text; it never computes the advertised metrics or associates reporting periods and attribution metadata with the figures. Generic guardrail text does not enforce those requirements, creating a report-integrity risk in a finance-facing workflow where incomplete or incomparable numbers could inform budget decisions. This is a reliability and integrity weakness, not evidence of malicious behavior or code execution.

Content

No source excerpt is available for this finding.