ClawHub
Back to skill

Security audit

TikTok Video Script Writer

Security checks for vulnerabilities and agentic risk

Overview

The bundled workflow uses limited local file access, though its helper only formats notes and does not automatically check TikTok limits.

Review the generated digest carefully: heading matching can omit template content, and caption length and disclosures are not automatically validated. Existing digest.md files in the chosen output directory are overwritten. Treat the optional unsigned Windows app as a separate software decision; this review does not establish its safety.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The main routine only matches input headings to predefined section names and copies their nonempty lines into a report. It contains no script-generation logic or platform-limit checks. Its primary behavior is therefore a notes digest formatter, materially narrower than the declared script-writing and validation purpose. Local input reads and output writes are ordinary supporting implementation details, not separate mismatches.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_write.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The manifest promises script writing measured against TikTok platform limits, but the implementation only reformats existing notes without generating script content or checking limits.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The run function groups existing Markdown content under matching headings and copies it into digest.md, inserting placeholders for missing sections. It neither generates a spoken hook or beat-by-beat script nor measures caption length or other platform limits, so its output does not provide the writing and validation described in the manifest.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.