Back to skill

Security audit

Emocity Biometric Scan

Security checks for vulnerabilities and agentic risk

Overview

This skill handles sensitive face and voice-derived readings, but its behavior is disclosed, user-initiated, purpose-aligned, and bounded to the stated scan/API workflows.

Install only if you are comfortable using emo.city and, optionally, RapidAPI for sensitive face/voice-derived readings. Use it only on yourself or consenting people, treat the output as casual and non-medical, and avoid optional file uploads unless you accept sending that media to the EmoPulse API with your own key.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill promotes camera, microphone, biometric, and affective inference features without a prominent up-front warning about the sensitivity of the data involved or the privacy implications before users are encouraged to start scanning. Because it normalizes collection of highly sensitive face/voice-derived attributes such as stress, heart-rate estimate, mood, micro-expressions, and voice tension, users may consent without understanding the scope, risks, or limits of the analysis.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The skill instructs the agent to poll an external service and retrieve biometric scan results tied to a one-time code. Even though the design claims only summary numbers are returned, this is still transmission and retrieval of sensitive biometric and inferred health/emotional data through a third-party endpoint, creating privacy and data-handling risk if the code is exposed, logs are retained, or the remote service is compromised.

Content

Scanner excerpt · SKILL.md (reported line 97)May include surrounding context.

text
   for i in $(seq 1 60); do
     R=$(curl -s --max-time 10 "https://api.emo.city/api/v1/agent-result/$CODE")
     case "$R" in *'"waiting"'*) sleep 5 ;; *) echo "$R"; break ;; esac
   done

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The manifest presents itself as a workout-readiness or fitness skill, but the description advertises biometric face scanning, heart-rate estimation, mood inference, microphone analysis, and file uploads. This mismatch can mislead reviewers and users about the true behavior of the skill, weakening informed consent and potentially allowing sensitive biometric collection under a less alarming identity.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The manifest describes collection and handling of highly sensitive inputs including camera video, microphone audio, and uploaded photos or clips, but does not provide an explicit privacy warning about what data is collected, where it is processed, whether it is transmitted to third parties, or how results are returned to the agent. In a biometric-analysis context, missing disclosure materially increases the risk of uninformed consent, privacy harm, and unsafe transmission of sensitive personal data.

Content

No source excerpt is available for this finding.