Back to skill

Security audit

quote-invoice-workbench

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local quote and invoice drafting helper with disclosed inputs, a small auditable calculator script, and no evidence of hidden network, credential, persistence, or destructive behavior.

Installers should treat this as a drafting aid, not an authoritative accounting or legal tool. Review rates, taxes, discounts, payment terms, and assumptions before sending any generated quote or invoice externally.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The skill promises quote, SOW, and invoice drafting capabilities that the detected implementation reportedly does not actually provide. This can mislead users into relying on incomplete or malformed financial outputs, which is risky in a billing and commercial context because omissions in assumptions, scope, or invoice content can cause disputes, underbilling, overbilling, or bad business decisions.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger examples include broad natural-language phrases such as 'invoice draft' and 'scope and quote' that could plausibly appear in ordinary conversation, making unintended skill activation more likely. In an agent environment, accidental invocation can cause the model to switch workflows, request files, or produce business documents when the user did not explicitly intend to use this skill.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

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.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill lists Chinese-only trigger phrases alongside English ones, and repeats Chinese suggested prompts, but nowhere states that language choice is optional or user-selectable. This can violate language/locale policy if the skill implicitly expects or steers users into a specific language without consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
73% confidence
Finding

The document presents major headings and usage guidance in Chinese while other operational content and examples are in English, but it does not state whether the skill should respond in Chinese, English, or follow user preference. This can amount to an implicit language/locale constraint without opt-in or explicit choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown file presents the skill’s goals, checks, and validation steps entirely in Chinese, which can impose a specific language on users or reviewers without indicating that another language is available. The policy allows locale or language constraints only when explicitly justified or offered as an opt-in choice, which is not present here.

Content

No source excerpt is available for this finding.