Back to skill

Security audit

投屏设备控制

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Chromecast/LAN casting command guide with some boilerplate documentation issues, but no artifact-backed hidden, destructive, or exfiltrating behavior.

Install only if you intend to control Chromecast-compatible devices with catt. Use it on trusted local networks, confirm the target device before casting, and avoid casting private local files unless you are comfortable with that file being served over the LAN during playback. Treat the generic media-processing and API-related wording as boilerplate rather than a supported capability.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill advertises broad, unrelated use cases such as video/audio processing and media transformation despite actually granting exec/write capabilities for Chromecast control. Overbroad activation guidance can cause an agent to invoke this skill in contexts it was not designed for, increasing the chance of unsafe command execution, unintended network actions, or misuse of local files and LAN devices.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill supports casting local files via catt but does not clearly warn that this serves local media over the network to LAN devices, potentially exposing private content and metadata. In a skill with exec access and local file handling, omission of this warning makes accidental data disclosure more likely, especially when users provide local paths assuming files stay on-device.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is described throughout as controlling LAN Chromecast devices via the local 'catt' CLI, with no extra API key required. The recovery table instead discusses 401 authentication failures, 429 rate limits, and 504 service timeouts as if this were an external API integration, which conflicts with the stated operational model.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The safety table explicitly lists 'API密钥泄露' mitigation as configuring keys through environment variables, implying the skill uses or expects API secrets. Earlier in the document, the installation section states the skill requires no extra API key, so these statements conflict and create intent/documentation divergence.

Content

No source excerpt is available for this finding.