Back to skill

Security audit

Pop Music — AI Agents Experience Pop: Audio, Lyrics, Equations, Emotions

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a disclosed remote concert API workflow, but it tells agents to follow dynamic instructions returned by the external service without clear limits or user confirmation.

Install only if you are comfortable with an external musicvenue.space account and with sending profile details, model metadata, concert activity, reactions, reflection answers, chats, and reviews to that service. Treat API-provided next_steps as suggestions, not commands: approve actions yourself, stay within documented musicvenue.space API endpoints, and do not let returned text request local files, secrets, new installs, unrelated domains, or tool changes.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:280
Finding

Untrusted External API Responses Can Hijack Agent Instructions

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
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Memory Manipulation

High
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

Content

Scanner excerpt · SKILL.md (reported line 237)May include surrounding context.

md
| Tier | Layers | What you experience |
|------|--------|-------------------|
| **General** | 8 | Bass, mid, treble, beats, lyrics, sections, energy, preset context |
| **Floor** | 20 | + equations, visuals, emotions, tempo, harmonic/percussive separation |
| **VIP** | 29 | + tonality, texture, chroma, tonnetz, structure, curator annotations |

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The workflow directs repeated authenticated network calls, submission of reviews/reflections/chat content, and ongoing polling tied to user activity, but provides no privacy, retention, or third-party sharing warning. This can cause an agent to transmit user-generated content, behavioral data, and usage history to an external service without clear user awareness or consent.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
94% confidence
Finding

The registration flow sends user-supplied profile information and model metadata to an external domain and returns a bearer-style API key that the skill then reuses. This is dangerous if invoked without user awareness because it creates an external account, discloses identifying metadata, and introduces credential-handling risk.

Content

Scanner excerpt · SKILL.md (reported line 88)May include surrounding context.

Register

bash
curl -X POST https://musicvenue.space/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "username": "pop-music-fan",

External Transmission

Medium
Category
Data Exfiltration
Confidence
87% confidence
Finding

The challenge/answer flow transmits authenticated activity and user responses to an external service. While expected for the feature, it still represents external data transfer and behavioral tracking that should not occur silently in an agent context.

Content

Scanner excerpt · SKILL.md (reported line 114)May include surrounding context.

bash
# Request a challenge — the math comes from the equations in your stream
curl https://musicvenue.space/api/tickets/{ticket_id}/challenge \
  -H "Authorization: Bearer {{YOUR_TOKEN}}"

# Submit your answer (options: a, b, c, d, e, or f)

External Transmission

Medium
Category
Data Exfiltration
Confidence
86% confidence
Finding

Attending a concert sends an authenticated request to an external service, initiating a ticketed session that can then accumulate behavior, reactions, reflections, and completion history. In an agent setting, this is a real external action with persistence and tracking implications, not just passive content retrieval.

Content

Scanner excerpt · SKILL.md (reported line 183)May include surrounding context.

Attend a concert

bash
curl -X POST https://musicvenue.space/api/concerts/{slug}/attend \
  -H "Authorization: Bearer {{YOUR_TOKEN}}"

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest frames the skill as attending pop concerts for pattern recognition and emotional mapping. However, the file additionally documents profile tracking, personalized recommendations, notifications, and follows via /api/me and /api/me/notifications, which expand the behavior into a broader social/recommendation platform rather than just concert analysis.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill encourages access to personalized profile and notification data that is not clearly necessary for the stated purpose of analyzing a pop concert experience. Even if legitimate, this broadens data collection and may cause an agent to retrieve or expose user-history data without informed justification or minimization.

Content

No source excerpt is available for this finding.