T09 · Insecure Skill Coding Practices
- Location
scripts/videogen.py:399- Finding
Untrusted API Fields Permit Download Path Traversal
- Content
View full analysis
- Remediation
View remediation
Security audit
Security checks for vulnerabilities and agentic risk
The skill is coherent for AI Hive Douyin video generation, but its downloader can write outside the chosen output folder if task metadata from the API is crafted maliciously.
Review before installing if you will use custom or untrusted AI Hive base URLs, because crafted task metadata could cause downloaded output files to be written outside the selected output directory. Use a virtual environment, avoid elevated privileges, keep the API key private, and only upload media you are comfortable sending to AI Hive.
scripts/videogen.py:399Untrusted API Fields Permit Download Path Traversal
SKILL.md:79Unpinned Third-Party Dependency Installation
This markdown file contains user-facing natural-language content exclusively in Chinese across the substantive changelog entries. Under the policy, forcing a specific language without user opt-in or justification is a natural-language policy violation.
The skill exposes shell, network, file read/write, and environment-dependent execution examples but does not declare any explicit tool scope or permissions boundaries. That makes it easier for an agent runtime to invoke the skill with broader-than-necessary capabilities, increasing the chance of unintended command execution, filesystem access, dependency installation, or network use if the skill is triggered in the wrong context.
The description uses broad 'Use this skill for ...' language across many marketing and video-production scenarios without tight trigger conditions. In agent environments, this can cause over-invocation or invocation in ambiguous contexts, which is risky here because the skill includes executable shell commands, package installation, file operations, and networked generation workflows.
The manifest describes a skill focused on generating and editing Douyin-oriented videos, including text-to-video, image-to-video, reference video, remix, and extension workflows. However, the same file implements a generic AI Hive chat/text client and a full chat command for multimodal text interaction, which is outside the stated video-editing purpose.
Beyond video generation, the code includes an image command for general image generation plus user-info and models commands for account/model introspection. Those are broader AI-platform utility functions rather than capabilities clearly implied by a Douyin viral-video generation/editing skill.
This code embeds natural-language strings and metadata that position the skill specifically for '抖音爆款短视频' generation and presents the CLI description/help text in Chinese only. The policy requires flagging language or locale constraints when they are imposed without explicit user choice or documented justification as a region-specific compliance tool.
The skill instructions and examples are predominantly in Chinese, and there is no natural-language indication that users may choose another language or locale. This can violate language/locale policy when a skill effectively forces one language without user opt-in.
The placeholder value 请替换为你的APIKey forces a specific language in user-facing configuration text. There is no indication that the skill is region-specific or that alternative locales are supported, so this appears to violate the language/locale policy for natural-language content.
The manifest describes content-generation and editing workflows, but this file also includes an interactive onboarding routine that launches a browser, walks the user through account login/API key creation, and writes credentials locally. That is an account-bootstrap capability, not a direct video-generation function.