Security audit

Hyperframes Media

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed media-generation helper whose credential use, cloud APIs, local commands, and background jobs fit its stated audio/video purpose.

Install only if you are comfortable with a media workflow that can use local credentials, cloud media providers, runtime package/model downloads, and a detached BGM render process. Run the preflight, choose cloud versus local generation deliberately, avoid untrusted parent-directory .env files, and do not use the OpenAI or Groq transcription examples with private audio unless you are allowed to upload that content.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 88% confidence
Finding: The skill documentation directs use of environment-backed credentials such as $HEYGEN_API_KEY, $HYPERFRAMES_API_KEY, $ELEVENLABS_API_KEY, and ~/.heygen, but the skill declares no permissions for env or secret access. That creates a transparency and least-privilege issue: an agent may access sensitive local credentials without the user understanding that the skill requires them.

Vague Triggers

Medium

Confidence: 76% confidence
Finding: The description is very broad and covers many common media tasks such as voiceover, TTS, BGM, SFX, transcription, captions, subtitles, and background removal. Over-broad routing can cause the skill to activate in situations the user did not specifically intend, increasing the chance of unnecessary command execution, credential use, or handling of local media files.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The documentation provides ready-to-run examples that upload local audio files to OpenAI and Groq transcription APIs, but it does not explicitly warn that doing so sends potentially sensitive media content to third-party services. In a media/transcription skill, users may process private recordings, so omission of a privacy/data-disclosure warning can lead to unintentional exposure of confidential audio and associated metadata.

Static analysis

Detected: suspicious.dangerous_exec, suspicious.env_credential_access

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/lib/bgm.mjs:31

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/lib/python.mjs:7

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/lib/tts.mjs:30

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/lib/tts.spawn.test.mjs:11

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: scripts/lib/heygen.mjs:33

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: scripts/lib/tts.mjs:28