Back to skill

Security audit

Subtitle QC

Security checks for vulnerabilities and agentic risk

Overview

This local subtitle tool needs review because its report options can overwrite the original file despite its preservation promise.

Keep a backup and use distinct, unused paths for the repaired subtitle and both reports, avoiding links to existing files. Prefer a version that validates every output path before writing. The scanner's partial-coverage exception was resolved by inspecting the output filename example and bundled code; the confirmed concern is accidental data loss.

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
  • 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 (2)

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 fix command can overwrite its input through --log or --todo despite its explicit non-overwrite guarantee.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The fix command opens --log and --todo paths in write mode without checking whether they identify the input subtitle, so either option can truncate the original and replace it with JSON. This contradicts the explicit preservation guarantee and can cause data loss through mistaken or adversarially supplied arguments; the recommended temporary-report workflow reduces exposure but does not enforce protection.

Content

No source excerpt is available for this finding.