Video Edit Compare

• Outcome: Review
• Latest audit: Sep 16, 2026 · 10:22 AM UTC
• Version: 1.0.1

Overview

This is a purpose-built video comparison skill that runs disclosed media-processing commands and writes disclosed review artifacts, with no evidence of hidden data theft or system persistence.

Install only in a workspace where you are comfortable letting ffmpeg read the selected source and final videos and write review files. Confirm the local video-understand/projectlib dependency is trusted, and prefer running it on explicit project paths rather than a sensitive general-purpose directory.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Undeclared Tool Scope

Medium

• Category: MCP Least Privilege
• Confidence: 94% confidence

Finding

The skill instructs the agent to invoke Python/ffmpeg and to write outputs under work/, review/, and cache paths, but it does not declare any explicit tool scope such as allowed-tools or permissions. That creates an authorization gap where an orchestrator may grant broader shell and file-write access than is actually needed, increasing the blast radius if the skill is modified, misused, or run in a sensitive workspace.

Content

No source excerpt is available for this finding.

subprocess module call

Medium

• Category: Dangerous Code Execution
• Confidence: 70% confidence

Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/make_compare.py (reported line 64)May include surrounding context.

python

def _probe(path):
    result = subprocess.run(
        [
            "ffprobe", "-v", "error", "-select_streams", "v:0",
            "-show_entries", "stream=width,height:format=duration", "-of", "json", str(path),

subprocess module call

Medium

• Category: Dangerous Code Execution
• Confidence: 70% confidence

Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/make_compare.py (reported line 83)May include surrounding context.

python

def _audio_hash(path):
    result = subprocess.run(
        [
            "ffmpeg", "-hide_banner", "-loglevel", "error", "-i", str(path),
            "-map", "0:a:0", "-c", "copy", "-f", "hash", "-hash", "md5", "-",

subprocess module call

Medium

• Category: Dangerous Code Execution
• Confidence: 70% confidence

Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/make_compare.py (reported line 95)May include surrounding context.

python

def _crop_bytes(path, at_s, x, y, size):
    sample_size = min(16, size)
    result = subprocess.run(
        [
            "ffmpeg", "-hide_banner", "-loglevel", "error", "-ss", f"{at_s:.6f}",
            "-i", str(path), "-vf",

subprocess module call

Medium

• Category: Dangerous Code Execution
• Confidence: 70% confidence

Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/make_compare.py (reported line 336)May include surrounding context.

python

if args.filter_only:
        return
    Path(args.output).parent.mkdir(parents=True, exist_ok=True)
    subprocess.run(command, check=True)
    output = Path(args.output).resolve()
    verify_output(timeline, args.source, args.final, output)
    parts = source_time_parts(timeline)

Missing User Warnings

Low

• Category: Not specified by scanner
• Confidence: 89% confidence

Finding

This Python file writes several artifacts to disk, including the filtergraph file, generated label images, a comparison summary markdown file, and updates project JSON metadata. While the script's purpose involves producing output, there is no confirmation prompt, visible warning comment/docstring, or user-facing disclosure in the file that it will also modify project state and create auxiliary files beyond the main output video.

Content

No source excerpt is available for this finding.

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical

• Code: suspicious.dynamic_code_execution
• Location: tests/test_make_compare.py:9