ClawHub
Back to skill

Security audit

AIOZ音频上传

Security checks for vulnerabilities and agentic risk

Overview

The skill is mainly an AIOZ audio uploader, but it asks for broad local and command access without tight limits around what files, credentials, and callback URLs may be used.

Review before installing. Use it only for audio files you intentionally want to upload to AIOZ Stream, avoid private or copyrighted media unless authorized, provide only the specific file path and callback URL you approve, and keep the AIOZ secret key out of logs and chat output. The publisher should narrow the tool permissions, remove unrelated invocation categories, and clarify the credential variables.

SkillSpector

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 (4)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill declares `exec`, `read`, and `write` capabilities even though its core function is uploading audio to a remote API. In an agent environment, unnecessary command execution materially expands the attack surface: a malformed prompt or future skill logic could access local files, exfiltrate secrets, or run shell commands unrelated to audio upload.

Intent-Code Divergence

Low
Confidence
81% confidence
Finding
The skill says authentication uses `stream-public-key` and `stream-secret-key`, but later shows a generic `API_KEY` environment variable configuration. This inconsistency can cause operators or agents to misconfigure secrets, increasing the chance that credentials are mishandled, logged incorrectly, or sent in the wrong place.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The description says the skill should be used for broad categories like video processing, audio editing, media conversion, and dubbing generation, which go well beyond the actual documented upload function. Overbroad invocation guidance can cause an agent to select this skill in unrelated contexts, unnecessarily granting it file and execution access and increasing the chance of unintended data transfer to the external AIOZ service.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs the agent to upload local audio files, send API keys in headers, and optionally transmit a `callback_url`, but it does not require a clear user-facing consent step or explicit warning that data will leave the local environment. In an agent setting, this creates a real risk of unintentional exfiltration of private media, embedded metadata, or sensitive callback endpoints to a third-party service.

Static analysis

No suspicious patterns detected.