ClawHub
Back to skill

Security audit

YouTube Data

Security checks for vulnerabilities and agentic risk

Overview

This YouTube research skill mostly matches its purpose, but it also exposes creator email lookup through a third-party API without prominent disclosure or controls.

Review this skill before installing if you do not want agents to retrieve creator email addresses or send channel/search targets to superagnt. For normal YouTube metadata, transcript, comment, and search workflows the external API use is expected, but email lookup should be used only with clear user intent and a legitimate purpose.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill does not clearly warn users that their queries, channel IDs, video IDs, and other inputs are sent to an external third-party API. This creates a transparency and privacy problem because sensitive research terms or user-provided targets may be transmitted off-platform without adequate notice.

External Transmission

Medium
Category
Data Exfiltration
Content
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
One call proves the key, the credit balance, and this source end to end:

```bash
curl -s https://api.superagnt.com/v1/credits \
  -H 'Authorization: Bearer $SUPERAGNT_API_KEY'
```
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill is presented as YouTube content research and channel intelligence, but it documents email-retrieval endpoints for channels. That expands the capability into contact discovery/PII acquisition without prominent disclosure, creating a mismatch between declared purpose and actual data access that could enable scraping or targeted outreach.

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
Channel email lookup is context-inappropriate for a skill marketed as YouTube research. Users may invoke the skill expecting metadata, comments, and transcripts, but instead gain access to creator contact data, increasing privacy and misuse risk beyond the stated use case.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill exposes email-enrichment endpoints without a corresponding warning that it can retrieve channel email addresses. This omission reduces informed consent and increases the risk of the tool being used for unsolicited contact harvesting or other privacy-invasive activity.

Static analysis

No suspicious patterns detected.