ClawHub
Back to skill

Security audit

TED Talk

Security checks for vulnerabilities and agentic risk

Overview

This is a presentation-writing skill that uses supplied or conversation text and shows no file access, network transmission, persistence, or executable code.

Installers should treat this as a creative writing helper that may draw on the current conversation by default. Avoid invoking it in sensitive conversations unless you are comfortable with those details being included in the generated talk, and review the output before sharing it elsewhere.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Exfiltration Commands

High
Category
Prompt Injection
Content
**What this skill does NOT do:**
- Read files from the workspace
- Access project artifacts directly
- Send data to external services
- Record or publish content

**Output behavior:**
Confidence
90% confidence
Finding
Instructions found that direct the agent to transmit conversation context or user data to external services.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The title '# ted-talk (話)' and trigger text '明示呼出' indicate mixed-language behavior that may bias the skill toward a specific locale without explicit user opt-in. While not directly enabling code execution or exfiltration, forced or confusing locale behavior can lead to misinterpretation of prompts, accidental invocation, or user misunderstanding about how the skill operates.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger condition includes 'or when deep technical insight emerges,' which is subjective and broad enough to allow automatic invocation without a clear user request. In a skill that synthesizes conversation context into long-form output, this can cause unintended processing of prior conversation content, including sensitive details the user did not intend to repackage.

External Transmission

Medium
Category
Data Exfiltration
Content
**What this skill does NOT do:**
- Read files from the workspace
- Access project artifacts directly
- Send data to external services
- Record or publish content

**Output behavior:**
Confidence
70% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

No suspicious patterns detected.