Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md [English](./SKILL.md) · [中文](./SKILL-cn.md)
Security audit
音色库搜索 ElevenLabs Voice Search
Security checks for vulnerabilities and agentic risk
Overview
The skill appears to be a third-party cloud CLI wrapper for ElevenLabs voice search, but its documentation also describes broader generation, file upload, local saving, and image-output behavior that does not clearly fit voice search.
Review this skill carefully before installing. It may be legitimate, but only use it if you are comfortable installing a third-party npm CLI, storing or passing a dLazy API key, and sending prompts or possibly local file references to dLazy-hosted services. Treat the documented image/generation/upload behavior as unresolved until the publisher clarifies the exact ElevenLabs search output and file-handling scope.
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
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- 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 (8)
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The skill is presented as a narrow ElevenLabs voice search utility, but the documentation describes generic prompt submission to a hosted inference API, async generation controls, local file uploads, and saving result assets. This broader-than-claimed behavior increases risk because users and agents may disclose prompts or local files under the assumption of a simple search operation, while the underlying workflow behaves like a general cloud generation client.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The documented output schema contradicts the stated purpose of the skill: a voice-library search tool should return voice metadata and audio preview references, but the example shows an image asset hosted on files.dlazy.com. This kind of mismatch is dangerous because agents may trust the declared behavior while actually handling unrelated remote assets, enabling misleading execution paths, unsafe assumptions about content type, and possible unintended downloads or downstream processing.
- Content
Whitespace Padding
Medium
- Category
- Prompt Injection
- Confidence
- 70% confidence
- Finding
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
- Content
md ## 错误处理 | Code | 错误类型 | 示例信息 | | ---- | -------------------------------- | ------------------------------------------------------------------------------------------------------------------------ | | 401 | 未授权 (API Key缺失或无效) | `ok: false, code: "unauthorized", message: "API key is missing or invalid"` | | 501 | 缺少必填参数 | `error: required option '--prompt <prompt>' not specified` |
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
The error documentation references image/video file reads and safety-policy generation failures, which do not fit the advertised ElevenLabs voice search workflow. Such contradictions are a security concern because they suggest copy-pasted or generic behavior from another tool, making it hard for users and agents to understand when local file access, remote generation, or policy enforcement may actually occur.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The skill is presented as a narrow ElevenLabs voice-search tool, but the documentation states that prompts are sent for inference and that local image/video/audio paths may be uploaded to remote storage. That scope expansion is security-relevant because it can mislead users and agents into providing local files or broader prompts to a tool they reasonably expect to perform only search, increasing the risk of unintended data exfiltration to third-party endpoints.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
The output schema shows an image result hosted on a remote files domain even though the skill is advertised as a voice-library search tool returning playable previews. This contradiction can cause agents or users to treat unexpected remote assets as normal output, weakening scrutiny around data handling and enabling misuse or deceptive repurposing of the skill's actual behavior.
- Content
Whitespace Padding
Medium
- Category
- Prompt Injection
- Confidence
- 70% confidence
- Finding
Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
- Content
md ## Error Handling | Code | Error Type | Example Message | | ---- | ---------------------------------- | ------------------------------------------------------------------------------------------------------------------------ | | 401 | Unauthorized (No API Key) | `ok: false, code: "unauthorized", message: "API key is missing or invalid"` | | 501 | Missing required parameter | `error: required option '--prompt <prompt>' not specified` |
Static analysis
No suspicious patterns detected.