Back to skill

Security audit

Excel公式工具

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be a spreadsheet helper, but it asks for broad file, write, and command-execution capability without enough scoping or user-control guidance.

Review this skill before installing. It may be useful for Excel formula help, but only use it in an environment where file writes and shell commands are restricted or require explicit approval, and avoid giving it sensitive spreadsheets unless you are comfortable with those broad capabilities.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Vague Triggers

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The activation condition is overly broad and covers generic file processing, document conversion, format transformation, and content extraction, far beyond an Excel-formula helper. Such broad triggering increases the chance the skill is invoked in unrelated contexts where its powerful tools may be misapplied, causing unintended access or modification of user data.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill declares exec capability even though generating and diagnosing Excel formulas does not inherently require arbitrary shell command execution. Granting exec in this context creates an unnecessary path to run system commands, access local resources, or chain with write capability for persistence or data exfiltration.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill is presented as a narrow Excel-formula assistant, but its description and operational guidance imply much broader file handling and command-execution behavior. This mismatch can mislead users and policy layers into granting dangerous capabilities under a benign-seeming label, increasing the chance of unexpected file, process, or network actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
80% confidence
Finding

The description explicitly says '支持中文交互' and the document is primarily written in Chinese without stating that users may choose another language. This can be read as a locale/language constraint that is not presented as an opt-in choice or justified as region-specific.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The documentation advertises execute and write capabilities without clearly warning that these can modify the local system or files. Users may invoke the skill assuming it only generates formulas, while the skill has the ability to make persistent or destructive changes if misused or prompted adversarially.

Content

No source excerpt is available for this finding.