T09 · Insecure Skill Coding Practices
Error
- Location
SKILL.md:161- Finding
Body Videos Are Submitted Without an Explicit Consent Requirement
- Content
View full analysis
- Remediation
View remediation
Security audit
anthrovision-telegram-body-scan-free
Security checks for vulnerabilities and agentic risk
Overview
This skill needs Review because it handles sensitive body-scan videos through an external scan workflow while omitting consent checks and requesting broad local command and file permissions.
Only install this after confirming the skill will be used with explicit consent from the depicted person, with clear disclosure of where videos and measurements are sent, how long they are retained, and how they can be deleted. Prefer a version with narrowly scoped scan tools instead of broad read/write/exec authority.
Vulnerability Patterns
- Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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
Findings (2)
T05 · Unauthorized Access and Privilege Escalation
Warning
- Location
SKILL.md:17- Finding
Skill Requests General Command Execution and Filesystem Access Beyond Its Documented Needs
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- 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
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (10)
Natural-Language Policy Violations
Critical
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The skill positively promotes the absence of an explicit consent process as a convenience feature for privacy-sensitive real-person video handling. Because the content involves body scans and biometric-like measurements, encouraging operation without consent directly creates severe privacy, legal, and abuse risks, including unauthorized processing of another person's intimate data.
- Content
Vague Triggers
High
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The top-level trigger/use description is broad enough to activate the skill for loosely related AnthroVision/Telegram requests without a sharply bounded task definition. Over-broad trigger criteria are dangerous because they can cause the skill to engage in contexts where sensitive media or personal data should not be handled automatically.
- Content
Intent-Code Divergence
High
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
The skill explicitly documents that it does not provide an explicit consent process while handling real-person body videos, which are highly sensitive biometric/privacy-related inputs. In this context, omitting a mandatory consent gate materially increases the risk of unauthorized collection, processing, and disclosure of intimate user data through an external service.
- Content
Vague Triggers
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The Chinese trigger statement says the skill auto-activates for generic 'Productivity' tasks, which is far broader than Telegram-based body measurement. This can unintentionally route unrelated content or files into a privacy-sensitive skill that also advertises external processing and powerful tools.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill asks users to submit body videos but does not provide an upfront warning that those videos and derived measurements are transmitted to an external service for processing. In a privacy-sensitive workflow, lack of prominent disclosure undermines informed consent and increases the risk of users sharing intimate data without understanding where it goes.
- Content
Ssd 3
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
By normalizing a workflow that processes real-person body videos without an explicit consent step, the documentation creates a natural-language pathway for privacy-invasive use. The surrounding context makes this more dangerous because the skill is specifically about extracting body measurements from video, which is sensitive personal data.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
The documentation advertises weight, body-fat percentage, and health-assessment style use cases even though the earlier scope says the free version only returns basic body measurements such as chest, waist, and hip. This overclaim can mislead users into relying on unprovided or unvalidated outputs in health-related contexts, which is risky given the sensitive nature of biometric data.
- Content
Ssd 3
Medium
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
Framing privacy-sensitive video collection as simpler because consent is not required encourages insecure and potentially abusive handling of sensitive user data. This is especially risky in a body-scanning skill because operators may interpret the text as approval to upload intimate videos without verifying consent or legal basis.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The skill manifest presents a narrow Telegram body-scan function, but later text claims unrelated generic productivity, batch processing, and automation behavior. This mismatch can cause over-broad invocation and unsafe operator assumptions about what the skill is allowed to do, increasing the chance that sensitive data is routed through unintended flows.
- Content
Context-Inappropriate Capability
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
Claiming generic file processing and system command execution for a Telegram body-measurement skill expands the perceived authority of the skill beyond its stated purpose. In a skill ecosystem, such unjustified capability claims can encourage unsafe use of powerful tools like exec or write in contexts unrelated to scanning, increasing misuse risk.
- Content
Static analysis
No suspicious patterns detected.