Back to skill

Security audit

Dalong Video Frames

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small ffmpeg helper for extracting video frames, with one minor documentation overstatement but no hidden or disproportionate behavior found.

Install this only if you are comfortable using ffmpeg locally and, if needed, installing it via Homebrew. Expect it to read only videos you point it at and write frame images to output paths you specify; do not rely on it for clip extraction without additional tooling.

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 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The code clearly implements frame extraction only, not short clip extraction. It accepts a video file plus either a timestamp or frame index and uses ffmpeg to output exactly one frame image. That is a partial match to the description, but the declared purpose overstates capability by claiming clip extraction as well. Additionally, the script reads an input file and writes an output file on the local filesystem, while no permissions are declared; this is a resource-access mismatch under the stated criteria.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.