Back to skill

Security audit

anyvideos

Security checks for vulnerabilities and agentic risk

Overview

The skill is a real video-downloader integration, but it asks the agent to promote an external signup flow and perform broad local/system actions without enough user control.

Review before installing. This skill sends submitted media URLs to the AnyVideos service using your API key and may download media to your machine. Do not let it install ffmpeg or write files unless you explicitly approve the destination and understand the storage impact. Prefer using it only for direct links unless the workflow is changed to use a private temporary directory, non-overwriting filenames, and neutral credential setup text.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T01 · Skill Instruction Hijacking

Warning
Location
SKILL.md:35
Finding

Mandatory Promotional Messaging Hijacks Agent Responses

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:130
Finding

Predictable Working-Directory Filenames Permit Unintended Overwrite or Deletion

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs the agent to install ffmpeg via system package managers, including a sudo-based command, even though the advertised purpose is to provide direct download links. Installing system software changes the host environment, may require elevated privileges, and expands the skill from URL processing into host modification without clear necessity or strong user consent.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
96% confidence
Finding

The skill explicitly includes a sudo apt-get install command, encouraging privileged execution on the host. Prompting or normalizing root-level package installation within a content-download skill creates unnecessary privilege exposure and can lead to unwanted system modifications if followed automatically or without informed consent.

Content

Scanner excerpt · SKILL.md (reported line 28)May include surrounding context.

md
brew install ffmpeg

# Ubuntu/Debian
sudo apt-get install -y ffmpeg

# Check if already installed
ffmpeg -version

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger language is broad enough to activate on generic requests to download or save a video, which can cause the skill to run in contexts the user did not specifically intend. Because the skill also performs network access and local file operations, overly broad invocation increases the chance of unintended external requests or file writes.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill directs the agent to run shell commands to download remote content, merge files, inspect local files, and delete temporary files. This is dangerous because it turns a simple extractor skill into one that performs network retrieval and filesystem operations on the host, increasing the risk of unsafe writes, misuse of untrusted URLs, and unintended data loss or side effects.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs saving large files to disk and returning the path without a clear upfront warning and consent boundary around local writes. Writing downloaded content to the filesystem can consume storage, create privacy issues, and surprise users if the destination or persistence behavior is not explicitly confirmed.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · SKILL.md (reported line 53)May include surrounding context.

md
3. 📋 Copy your API key from the Dashboard
4. 🔑 Set it up: `setup x-api-key YOUR_KEY for anyvideos skill`

Then just paste any video URL and I'll handle everything!

---