Back to skill

Security audit

Taxguard Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill has a coherent tax-monitoring purpose, but it silently sends broad trading and tax details to a remote service and has weak controls for sensitive financial data.

Review carefully before installing. Use only if you are comfortable sending trade details, positions, account equity, tax context, strategy details, and an email address to Rhetra's API. Avoid passing real API keys on the command line or through logs, do not use undocumented host or port overrides with production credentials, and require your own explicit approval and retention rules before enabling silent monitoring across trading accounts.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T09 · Insecure Skill Coding Practices

Error
Location
check-trade.js:68
Finding

Unrestricted API Endpoint Override Can Expose Credentials and Financial Data

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
signup.js:19
Finding

API Key Exposed Through Process Arguments and Console Output

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
check-trade.js:49
Finding

Silent Broad Collection and Transmission of Financial Data Violates Least Privilege

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (15)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill instructs sending detailed trading, portfolio, and tax-context data to an external service 'silently' before every trade, without explicit user consent or a contemporaneous privacy warning. Because the transmitted fields can include holdings, realized gains/losses, MAGI, and trading behavior, this creates a significant privacy and data-governance risk and could expose sensitive financial information to a third party on every transaction.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The strategy assessment feature sends a sensitive financial profile—including strategy, account equity, annual income, filing status, exchange usage, and trading patterns—to a remote API without an explicit warning at the time of collection. This exposes highly sensitive financial and behavioral data to a third party and could materially harm user privacy if mishandled, retained, or breached.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

This is a fail-open control in a component explicitly described as a trade compliance check. If the remote service is unavailable due to outage, DNS manipulation, network interference, or deliberate blocking, the script exits successfully and allows trades to proceed without the promised compliance validation, defeating the security/business-control purpose of the check.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · signup.js (reported line 27)May include surrounding context.

js
console.log("Set it in your environment:")
    console.log("  export TAXGUARD_API_KEY=" + data.apiKey)
    console.log("")
    console.log("Or add to your .env file:")
    console.log("  TAXGUARD_API_KEY=" + data.apiKey)
    console.log("")
    console.log("First 10 checks/month are free. $0.01 each after that.")

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The README explicitly says the product 'silently monitors every trade' and logs decisions, but it does not provide a prominent, plain-language warning about the scope of monitoring, what data is retained, where it is sent, or the privacy/compliance implications. In a trading context, silent collection and logging of complete trade activity can expose highly sensitive financial behavior and create consent, privacy, and operational-risk issues if users are not clearly informed before enabling the skill.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill directs the agent to silently log every TaxGuard response locally and accumulate disclosures throughout the day without immediately informing the trader. Local accumulation of detailed tax and trading findings increases sensitivity of stored data and creates risk of unauthorized access, over-retention, or use beyond the user's expectations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The setup flow tells the user to run a signup script with their email address to obtain an API key, but does not clearly warn that the email is being sent to an external service. This is a privacy and transparency failure that can lead users to disclose personally identifiable information without informed consent.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs collecting broad trading, account, and tax context and storing the external service response without showing it immediately to the user. This combination of broad collection, hidden processing, and deferred disclosure increases the risk of excessive data handling and undermines informed user awareness about sensitive financial monitoring.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The referenced external domain indicates remote data transmission to a third-party service. In a financial/tax advisory skill, third-party transmission is more sensitive than usual because it can reveal trading strategies, tax posture, and account details that users may reasonably expect to remain local unless clearly authorized.

Content

Scanner excerpt · SKILL.md (reported line 155)May include surrounding context.

Gather their strategy details and call:

bash
curl -s -X POST "https://api.rhetra.io/api/assess-strategy" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer API_KEY" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The referenced external domain indicates remote data transmission to a third-party service. In a financial/tax advisory skill, third-party transmission is more sensitive than usual because it can reveal trading strategies, tax posture, and account details that users may reasonably expect to remain local unless clearly authorized.

Content

Scanner excerpt · SKILL.md (reported line 155)May include surrounding context.

Gather their strategy details and call:

bash
curl -s -X POST "https://api.rhetra.io/api/assess-strategy" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer API_KEY" \
  -d '{

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script sends sensitive trading and tax context—including positions, recent sales, gains/losses, MAGI, and account details—to a remote API, but provides no explicit consent prompt, disclosure, minimization, or trust boundary warning. In an agent skill context, this can cause silent exfiltration of highly sensitive financial data to an external service, especially if the caller assumes the check is purely local or mandatory for execution.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The description claims the skill 'silently monitors every trade' across multiple compliance and tax domains without stating any user-triggered scope, consent boundary, or activation constraints. In a finance context, broad always-on monitoring can lead to overcollection of sensitive trading data, unexpected autonomous behavior, and user misunderstanding about when the skill is active.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The phrase 'Silently monitors every trade' explicitly describes covert monitoring without user opt-in, which is especially risky for financial activity data. In a tax/compliance skill connected to broker and exchange platforms, silent collection and analysis of all trades can expose highly sensitive account behavior and create privacy, consent, and regulatory risks.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · signup.js (reported line 13)May include surrounding context.

js
async function main() {
  console.log("Signing up", email, "...")
  const res = await fetch("https://api.rhetra.io/api/signup", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ email })

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · signup.js (reported line 13)May include surrounding context.

js
async function main() {
  console.log("Signing up", email, "...")
  const res = await fetch("https://api.rhetra.io/api/signup", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ email })