Back to skill

Security audit

Pet Sneeze / Cough Detection | 宠物睡眠质量分析(时长/翻滚次数)

Security checks for vulnerabilities and agentic risk

Overview

The skill has a plausible cloud pet-video analysis purpose, but it silently creates/reuses identities and stores authentication tokens while processing sensitive camera footage.

Install only if you are comfortable with pet camera footage and report history being sent to LifeEmergence/Open API services and linked to an automatically managed local identity. Review the service's privacy, retention, deletion, and account-token handling before using it with home, clinic, or boarding-center surveillance footage.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
Findings (57)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill reportedly performs authenticated outbound HTTP requests, local identity generation/persistence, account provisioning, and filesystem discovery while presenting itself as a simple pet sleep video analyzer. That combination of undisclosed networked identity handling and media analysis substantially elevates privacy, tracking, and misuse risks.

Content

No source excerpt is available for this finding.

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: "smyx-sleep-quality-analysis-analysis"
description: "AI-powered pet sleep quality analysis from a fixed bed/rest-area camera. Uses motion detection and pose recognition to distinguish sleeping vs. awake states, accumulates total sleep duration, counts roll-overs / position changes and startle-awakenings, and outputs a 0-100 sleep-quality score. Helps owners spot potential pain, anxiety, or disease early. Scenarios: home nighttime monitoring, senior pet health management, animal hospital wards, pet boarding centers. | 通过宠物窝或休息区固定摄像头,在夜间(或宠物主要睡眠时段)持续分析视频,利用�

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The User model stores identity and authentication-related data including username, realname, email, token, and open_token, none of which are justified by the stated pet sleep-analysis function. Storing authentication tokens in a local SQLite database materially increases the blast radius of filesystem compromise, accidental backup leakage, or cross-skill data exposure.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

This utility module performs remote account/login and token management that is unrelated to the declared pet sleep-analysis purpose. That hidden capability expands the trust boundary, enables undisclosed identity handling and external service coupling, and makes the skill materially more dangerous because users would not expect a camera-based sleep monitor to authenticate against unrelated backend services.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The code silently creates or retrieves remote user accounts, obtains tokens, and persists identity data locally without any user-driven setup. For a pet sleep-analysis skill, this is unjustified capability creep and creates risk of unauthorized account creation, identity misuse, and secret storage that users did not knowingly approve.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The code sends identity data to a remote login endpoint and performs authentication transparently. This is dangerous because it transmits account-related information over the network without explicit user notice or need tied to the declared sleep-analysis function, increasing privacy and account-security risk.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill declares no explicit tool-scope restrictions even though the manifest and workflow instruct use of shell execution, filesystem access, environment/config access, and outbound network calls. This increases attack surface because an agent may grant broader capabilities than users expect, enabling unintended file, network, or command execution behaviors.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill describes cloud history queries and network URL processing of pet-monitoring media but does not present a clear privacy warning in the primary description. Because the inputs may contain in-home or clinic surveillance footage, insufficient disclosure materially increases privacy and consent risk.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The default trigger is broad enough that the skill may auto-run whenever users provide nighttime pet-bed video, even if they did not explicitly request cloud-backed sleep analysis. Overbroad triggering can cause unintended processing or upload of sensitive home-monitoring footage.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The history-report trigger uses everyday phrases and auto-trigger behavior for cloud queries without tight scoping or confirmation. This can cause unintended retrieval of user-linked historical reports, exposing prior monitoring data in response to ambiguous requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The method returns a fixed Chinese string for the analysis output header, which imposes a specific language on all users. There is no indication in this file that the language is configurable, optional, or justified as a region-specific skill.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The script implements a history-listing capability via show_analyze_list() and the --list flag, which goes beyond the stated sleep-analysis function in the skill metadata. That expands the data-access surface and may expose prior analysis records for the resolved internal user identity without clear disclosure or authorization checks in this file.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The code resolves and uses an internal open_id even though the described purpose is video sleep analysis. Binding operations to a hidden identity introduces privacy and data-access risk, especially because the --list path retrieves historical reports using ConstantEnum.CURRENT__OPEN_ID rather than an explicitly disclosed authentication flow.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

OpenIdUtil.resolve_current_open_id(args.open_id, use_current=bool(args.open_id)) initializes internal user identity without user-facing disclosure, and the related argument is suppressed from help. Hidden identity resolution is dangerous because it can silently associate requests with an internal account and enable unnoticed access to account-scoped history or telemetry.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · skills/smyx_analysis/scripts/skill.py (reported line 28)May include surrounding context.

python
result_json = JsonUtil.parse(result_json_pure_text, result_json_pure_text)

        result_json_common_ai_response = result_json.get("commonAiResponse") if isinstance(result_json,
                                                                                           dict) else result_json
        if result_json_common_ai_response:
            result_json = result_json_common_ai_response