Back to skill

Security audit

Temu 电商视频生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches its Temu video purpose, but it can send a stored AI Hive API key to any configured API URL, so endpoint settings need careful review.

Install only if you trust the AI Hive service and verify that the configured base URL remains the official endpoint. Do not use --base-url, AI_HIVE_BASE_URL, or config-file base_url values from an untrusted source; rotate the API key if it may have been used with an untrusted endpoint. Use an isolated environment and pinned dependency versions where possible.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/videogen.py:91
Finding

Stored API Key Can Be Sent to an Arbitrary API Endpoint

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:65
Finding

Unpinned Runtime Dependency Installation

Content
View full analysis
``` ### Technical Analysis The runtime instructions install `requests` without a fixed version, integrity hash, lock file, or isolated environment. The package version and its dependency graph are therefore selected from the configured package index at installation time. Although `requests` is a legitimate and widely used dependency, installing an unconstrained latest version makes the effective dependency set change after the Skill has been audited. A compromised package release, package repository, mirror, dependency, or local package-index configuration could introduce code that was not part of the reviewed project. Python packages can execute code during installation or when imported. The script imports `requests` immediately at startup, so a compromised installed package can also execute with the invoking user's privileges. ### Attack Path 1. A user follows the documented `pip3 install requests` instruction. 2. The user's package index, mirror, network path, or a future package release supplies a compromised package or transitive dependency. 3. `pip` downloads and installs the unreviewed version because no version or hash constraint is present. 4. Malicious code executes during package installation or when `scripts/videogen.py` imports `requests`. 5. The code runs with the privileges of the user performing the installation or invoking the Skill. This finding does not establish that the current `requests` package is malicious. The risk arises from mutable and unverified dependency resolution. ### Impact Assessment If the dependency supply chain is compromised, attacke ...[truncated 650 chars]
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (11)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill advertises and invokes shell execution, package installation, network access, and file operations, but it does not declare any explicit tool scope or permission boundaries. This increases the chance that an agent will run the skill with broader privileges than necessary, enabling unintended filesystem or network actions if the script or its inputs are abused.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The description explicitly targets Chinese-language use ('Temu商品视频...') and multi-country localization support in a fixed way, rather than offering a user-selectable language/locale preference. Under the policy, language or locale constraints should be optional, user-chosen, or clearly justified as region-specific compliance tooling.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The prompt instructs the skill to 'Create a Japanese-market visual base' as a default example, which is a locale-specific constraint expressed in natural language. Because the file does not present this as a user-selected option or explain a compliance-driven need, it conflicts with the language/locale policy guidance.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file-level docstring says this script is a '通用 AI 能力调用工具' that wraps text chat, image generation, video generation, model querying, media upload, and task polling. That is materially broader than the manifest's Temu-specific product-video generation/editing purpose, indicating the implementation is a repurposed generic AI Hive client rather than a narrowly scoped Temu video skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This Python file contains user-facing natural-language strings and documentation entirely in Chinese, starting from the module docstring and continuing throughout prompts and help text. Because the skill exposes CLI guidance and interactive flows to end users without offering locale selection or opt-in, it imposes a specific language on users, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The code implements a full text chat endpoint and corresponding CLI flow for arbitrary chat interactions, including optional multimodal image input. A Temu ecommerce video generation/editing skill does not inherently require a standalone general chat capability, and the manifest does not declare conversational AI as part of the skill's scope.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The implementation includes a distinct image generation API operation and command path, allowing users to generate images independently of video workflows. The manifest is specifically about creating and editing Temu product videos and related video variants/localizations; standalone image generation is not justified by that stated purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown file contains user-facing natural language only in Chinese across the substantive changelog entries. Under the policy, forcing a specific language without user opt-in or a documented regional justification is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The example API key value includes Chinese text ('请替换为你的APIKey'), which imposes a specific language in user-facing configuration content. There is no indication that the skill is region-specific or that users may choose another language, so this appears to violate the language/locale policy criterion.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The code provides a user-info endpoint and command explicitly described as querying user information and wallet balance. Billing/account inspection is not an obvious requirement for a Temu ecommerce video generation/editing skill and is not mentioned in the manifest's purpose.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The script can list all available TEXT, IMAGE, and VIDEO models from AI Hive, which is a platform-exploration capability rather than a direct Temu video editing function. While some model lookup is needed internally, exposing generic model enumeration to the user is broader than the manifest's stated business purpose.

Content

No source excerpt is available for this finding.