Back to skill

Security audit

Skillboss

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward external AI gateway wrapper, but users should understand that prompts and the SkillBoss API key are sent to a remote service.

Install only if you are comfortable sending prompts, task inputs, and your SkillBoss API key to the SkillBoss/HeyBoss remote API and potentially downstream model providers. Avoid submitting secrets, regulated data, or proprietary content unless your organization permits that use.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (13)

Lp1

High
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The script makes multiple HTTPS requests to an external API but the static analysis indicates network capability is undeclared. Undeclared egress is dangerous because it can transmit user data and secrets outside the platform's expected trust boundary without an accurate permission prompt or review surface.

Content

No source excerpt is available for this finding.

Lp1

High
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The script makes multiple HTTPS requests to an external API but the static analysis indicates network capability is undeclared. Undeclared egress is dangerous because it can transmit user data and secrets outside the platform's expected trust boundary without an accurate permission prompt or review surface.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill routes user prompts and generated content to an external multi-provider AI gateway, but the description and usage guidance do not clearly warn users that their data will leave the local environment and be sent to third-party services. This creates a real privacy and data-handling risk because users may submit sensitive prompts, files, or generated outputs under the mistaken assumption that processing is local or confined to a single trusted provider.

Content

No source excerpt is available for this finding.

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 · scripts/run.mjs (reported line 29)May include surrounding context.

js
const body = { api_key: apiKey };
  if (args[1]) body.types = args[1];

  const resp = await fetch("https://api.heybossai.com/v1/models", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(body),

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · scripts/run.mjs (reported line 29)May include surrounding context.

js
const body = { api_key: apiKey };
  if (args[1]) body.types = args[1];

  const resp = await fetch("https://api.heybossai.com/v1/models", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(body),

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 · scripts/run.mjs (reported line 113)May include surrounding context.

js
const body = { api_key: apiKey };
  if (args[1]) body.types = args[1];

  const resp = await fetch("https://api.heybossai.com/v1/models", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify(body),

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This call transmits arbitrary user prompt content and the API key to a third-party model gateway. In the context of an AI gateway skill this behavior is core functionality, but it is still security-relevant because sensitive prompts may be exfiltrated to an external processor and the code provides no in-band notice, filtering, or data minimization.

Content

Scanner excerpt · scripts/run.mjs (reported line 42)May include surrounding context.

js
console.log(JSON.stringify(await resp.json(), null, 2));

} else if (cmd === "run" && args[1] && args[2]) {
  const resp = await fetch("https://api.heybossai.com/v1/run", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This call transmits arbitrary user prompt content and the API key to a third-party model gateway. In the context of an AI gateway skill this behavior is core functionality, but it is still security-relevant because sensitive prompts may be exfiltrated to an external processor and the code provides no in-band notice, filtering, or data minimization.

Content

Scanner excerpt · scripts/run.mjs (reported line 42)May include surrounding context.

js
console.log(JSON.stringify(await resp.json(), null, 2));

} else if (cmd === "run" && args[1] && args[2]) {
  const resp = await fetch("https://api.heybossai.com/v1/run", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

User-provided prompt content is sent directly to a third-party API in the run command without any user-facing disclosure in the script. This is a privacy and data-governance risk because users may paste secrets, proprietary data, or personal information, not realizing it will leave the local environment and be processed by an external provider.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This call sends task type and user-supplied inputs to an external auto-routing endpoint. The skill context makes outbound transmission expected, but the auto-selection design can obscure where user data goes, so the primary risk is undisclosed off-device processing of potentially sensitive content.

Content

Scanner excerpt · scripts/run.mjs (reported line 86)May include surrounding context.

js
? { messages: [{ role: "user", content: args[2] }] }
    : { prompt: args[2] };

  const resp = await fetch("https://api.heybossai.com/v1/pilot", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ api_key: apiKey, type, inputs }),

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This call sends task type and user-supplied inputs to an external auto-routing endpoint. The skill context makes outbound transmission expected, but the auto-selection design can obscure where user data goes, so the primary risk is undisclosed off-device processing of potentially sensitive content.

Content

Scanner excerpt · scripts/run.mjs (reported line 86)May include surrounding context.

js
? { messages: [{ role: "user", content: args[2] }] }
    : { prompt: args[2] };

  const resp = await fetch("https://api.heybossai.com/v1/pilot", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ api_key: apiKey, type, inputs }),

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The task command forwards prompt content to an external API without a clear disclosure to the user at the point of use. Because this command auto-selects models, the trust boundary is even less obvious, increasing the chance that sensitive prompts are sent externally without informed consent.

Content

No source excerpt is available for this finding.

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 · scripts/run.mjs (reported line 113)May include surrounding context.

js
}

} else if (cmd === "tasks") {
  const resp = await fetch("https://api.heybossai.com/v1/pilot", {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ api_key: apiKey, discover: true }),