Security audit

Budget Cash Flow Cleanup Helper

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only budgeting helper; its triggers are broader than ideal, but it does not install code, run commands, persist data, use credentials, or send data elsewhere.

Install this only if you want an agent to help with budgeting, transaction cleanup, subscription review, or cash-flow planning. Because the activation wording is broad, review when it triggers and share only the financial records needed for the task.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (11)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger sentence is overly broad and malformed, making activation criteria unclear and increasing the chance that the skill is invoked for loosely related requests. In an agent environment, ambiguous triggers can cause overreach, where the skill handles sensitive budgeting or transaction data when a more appropriate or narrower skill should have been used.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The invocation guidance does not define when the skill should activate versus when it should not, which can lead to accidental routing of unrelated financial or operational requests into this skill. Because the skill concerns bank exports and cash-flow analysis, misactivation may expose sensitive financial data to unnecessary processing or produce incorrect task handling.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger phrases are broad and loosely scoped, which can cause the skill to activate for generic finance or workflow requests that were not meant for it. In an agent environment, overbroad invocation increases the chance of inappropriate routing, unintended processing of sensitive financial data, or prompt-space collisions with other skills.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The manifest description uses very broad activation terms such as business-and-operations, budget, cash flow, bank export, and transaction categorization without meaningful scoping. This can cause the skill to activate for many ordinary finance-related requests, increasing the chance of inappropriate routing, unintended handling of sensitive financial data, and interference with more suitable skills.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger section defines broad keywords and example phrases with little boundary control, so common requests about budgets or cash flow may invoke the skill even when the user did not ask for transaction-cleanup help. In a finance context, over-triggering is more concerning because it can pull sensitive bank or card-export discussions into an unnecessarily broad skill workflow.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger phrases are broad, generic business terms such as 'budget', 'cash flow', and 'bank export', which can overlap with ordinary conversation and cause the skill to activate when the user did not intend to invoke it. This is primarily a security and safety boundary issue because accidental activation can route user input into the wrong workflow, increasing the chance of inappropriate data handling or misleading assistance.

Natural-Language Policy Violations

Medium

Confidence: 74% confidence
Finding: The file is a Chinese-language skill variant, but the documentation does not indicate any user-choice, locale negotiation, or bounded applicability for when this language-specific version should be used. In practice this can cause user misunderstanding, incorrect task execution, or accidental disclosure of sensitive financial information if the user cannot fully understand prompts, assumptions, or outputs.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The default prompt uses highly generic phrasing such as 'help me' plus broad business and budgeting concepts, which can overlap with ordinary user requests and cause the skill to activate when the user did not clearly intend it. This increases the chance of unintended tool routing, exposing user financial context to a skill unnecessarily or causing the model to follow this skill's workflow instead of the user's preferred path.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: Enabling implicit invocation without clear trigger constraints allows the system to auto-select this skill based on loosely related financial language. In a finance-oriented skill, that is more sensitive because users may discuss budgets, expenses, or business operations casually, leading to accidental activation and unnecessary handling of potentially sensitive financial data.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger sentence is so broad and natural-language-like that it can activate on generic user requests unrelated to a deliberate invocation of this skill. In an agent routing context, ambiguous activation boundaries can cause the skill to ingest sensitive financial data or steer workflows when the user did not explicitly intend to use this capability, increasing the chance of misrouting and privacy-impacting overreach.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: This trigger lacks clear boundaries for when the skill should or should not activate, making it easy for unrelated business, budgeting, or planning requests to match accidentally. Because the skill targets financial-record processing, accidental invocation can expose users to unnecessary collection or processing of sensitive transaction data and lead to incorrect automated assistance.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal