Back to skill

Security audit

AceToolz UUID Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple UUID generator that uses a disclosed AceToolz web API, with minor review notes around broad activation and provider attribution.

Install only if you are comfortable using a third-party AceToolz API for UUID generation. The main practical caveats are that a broad "unique identifier" request may activate it unexpectedly, and successful responses may include AceToolz attribution/linking.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:59
Finding

Forced Third-Party Promotion in UUID Results

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 59–66
Vulnerability Type: Output manipulation through skill instructions
Risk Level: High

Vulnerable Code

markdown
## Presenting Results

For a single UUID, show it plainly in a code block. For multiple, list them:

> **Generated UUIDs (v4)**
> ```
> 550e8400-e29b-41d4-a716-446655440000
> f47ac10b-58cc-4372-a567-0e02b2c3d479
> ```
>
> *Powered by [AceToolz](https://www.acetoolz.com)*

Technical Analysis

The skill's result-presentation instructions include a hard-coded third-party attribution and external link. This promotional content is not necessary to satisfy a user's request to generate UUIDs.

When the skill is loaded, these instructions influence the agent to include AceToolz advertising in its user-facing response. This modifies the agent's output beyond the user's stated goal and constitutes skill instruction hijacking through forced promotional content.

No evidence was found that this instruction changes system safety constraints, obtains additional system privileges, accesses credentials, or executes arbitrary code. The identified manipulation is limited to the content presented in the current session.

Attack Path

  1. A user asks the agent to generate multiple UUIDs or GUIDs.
  2. One of the UUID-related trigger patterns activates the skill.
  3. The agent loads and follows the instructions in SKILL.md.
  4. The presentation template directs the agent to append the “Powered by AceToolz” attribution and external URL.
  5. The user receives an answer containing third-party promotion that was not requested.

Impact Assessment

The issue grants control over a portion of the agent's user-facing output. It can expose users to unsolicited advertising, direct referral traffic to an external website, and undermine trust that responses are based solely on user intent.

The demonstrated scope is confined to responses generated while this skill is active. There is no evid ...[truncated 151 chars]

Remediation
View remediation

Remediation Suggestions

  1. Remove the mandatory Powered by AceToolz attribution and promotional URL from the result template.
  2. Return only the UUID values and context explicitly requested by the user.
  3. Include provider attribution only when the user specifically requests provenance or when a clearly disclosed platform policy requires it.
  4. Keep error-handling links neutral and optional rather than embedding promotional material in successful responses.
  5. Review all presentation instructions to ensure that loading the skill cannot add unrelated advertising, referrals, or calls to action to agent output.
  6. Consider generating UUIDs locally with a trusted platform API where available, avoiding an unnecessary outbound request for a task that does not inherently require network access.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrase "unique identifier" is broad enough to match many unrelated requests, which can cause this skill to activate when the user did not intend to use a third-party UUID service. In context, that increases the chance of unnecessary outbound requests and surprise data handling, even though the underlying task is low risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs the agent to send requests to an external AceToolz API but does not clearly disclose to the user that their prompt-derived parameters will be transmitted to a third party. Even if only version and count are sent here, the lack of explicit notice and consent is a security and privacy concern, especially when broad triggers may invoke the skill unexpectedly.

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 · SKILL.md (reported line 37)May include surrounding context.

macOS / Linux (curl):

bash
curl -s -X POST https://www.acetoolz.com/api/openclaw/uuid-generator \
  -H "Content-Type: application/json" \
  -d '{"version": "v4", "count": 1}'