Security audit

短剧视频、时间线与音乐提示词

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Chinese short-drama prompt-writing workflow with local validation helpers and no evidence of hidden execution, persistence, external service calls, or data exfiltration.

Install this only if you want a Chinese-oriented short-drama video prompt workflow. Before use, be aware it may read project storyboards, keyframes, and JSONL specs to produce or check prompt artifacts, and its helper scripts are local validators rather than media-generation tools.

A.I.G

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

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (27)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 99% confidence
Finding: The declared description presents a creative writing/prompt-formatting skill for short-drama video generation workflows. The supplied code instead implements a CLI audit tool for container/shot consistency checks: loading JSONL records, validating membership/order/durations, reconciling totals, and outputting machine-readable findings. This is a materially different primary purpose and capability set. There is no evidence of storyboard-to-prompt conversion, Markdown generation, music prompt writing, or any related text-authoring behavior.
Content: No source excerpt is available for this finding.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 99% confidence
Finding: 描述声称这是一个用于把分镜转换成视频提示词/配乐提示词的写作型技能，但代码没有任何提示词生成、Markdown 组织、配乐意图撰写、分镜文本改写相关逻辑。相反，代码的核心功能是静态校验：解析 JSONL 记录，解析显式时间区间，核对镜头接受时长，检测 overflow、shortfall、未声明重叠、声明总时长不匹配、引用失效等问题，并以 JSON 形式输出 findings。其主要目的与声明完全不同，属于实质性能力和用途不一致。虽然代码“读取文件、不写回源文件”这一点本身不危险，但这不改变其主功能是验证器而非提示词生成器，因此应判定为描述与行为不匹配。
Content: No source excerpt is available for this finding.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 99% confidence
Finding: 声明的核心用途是内容生成/改写：把分镜转成视频提示词，或撰写配乐/主题曲意图文本。实际代码没有任何分镜解析、Markdown 生成、视频提示词生成、配乐提示词撰写或主题曲意图创作逻辑。相反，它是一个严格的音乐规格校验脚本，验证 JSONL 记录格式、引用解析、字段白名单、时间范围、模式与歌词一致性、混音意图和状态值等。这不是对声明功能的支撑实现，而是完全不同的主要用途。因此属于明显的描述与行为不匹配。
Content: No source excerpt is available for this finding.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 96% confidence
Finding: The declared purpose is user-facing prompt authoring for video and music intent prompts based on storyboards/keyframes. The actual code does not implement prompt generation or storyboard transformation. Instead, it is a developer-oriented self-test harness for validating music specification records and error handling. While music-related validation is loosely adjacent to the declared domain, the primary behavior is materially different: schema testing and record validation rather than prompt writing. Therefore this is a clear description-behavior mismatch.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 70% confidence
Finding: Without declared permissions the skill's intent is opaque and cannot be validated.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The display name, short description, and default prompt are all specified only in Chinese, which indicates a fixed language choice in user-facing natural language. The file does not offer an opt-in, alternative locale, or justification that this skill is intended only for a Chinese-speaking or region-specific context.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The entire skill file is written in Chinese and includes no indication that another language is supported or that the locale is optional. Under the policy, a skill that effectively requires a specific language without user opt-in can be a natural-language policy violation unless the regional constraint is explicitly justified.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The JSON values include hardcoded Chinese locale/path naming such as "剧集/..." with no indication that language choice is optional or limited to a documented region-specific context. Under the policy, fixed language/locale usage in natural-language or config values can be a violation when no opt-in or justification is provided.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: The example source artifact path includes non-English locale-specific content (剧集/EP001/screenplay.md), indicating the skill may be oriented to a specific language context. The file does not state that language choice is optional or justified as a region-specific tool, so this can violate the policy against forcing a language/locale without user opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 81% confidence
Finding: This file contains natural-language guidance that centers execution on a specific language context ('方言文档') and the overall document itself is Chinese-only. Under the policy, forcing a language or locale without offering user choice can be a policy violation unless clearly justified, and no such opt-in or justification is present here.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 80% confidence
Finding: These lines state that if the spoken language is not predefined, it must be confirmed separately because it is independent from the language used for descriptive text. Across the document, the delivery format and instructions are written as Chinese-only norms, which effectively imposes a language choice for the description text without an explicit opt-in or stated regional justification.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The example profile hard-codes video_prompt_language: "en", and the document later reinforces that structural fields and section names must be in English. This is a natural-language locale constraint presented as a requirement, but the file does not offer a user choice or clearly justify the restriction as a region-specific compliance need.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 86% confidence
Finding: The line states that H3 structural fields must use English, which is a language policy constraint embedded in natural-language instructions. Because the file does not present this as an opt-in choice and does not narrowly frame it as an unavoidable interface limitation, it can conflict with the policy against forcing a specific language without user opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: This instruction explicitly requires both section names and body descriptions to use English. That is a direct locale/language mandate in the skill guidance, and the file does not offer a language choice or sufficiently constrain the rule to unavoidable system syntax only.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: L178要求为每条台词标注“口语语言”，L183进一步规定“非画内配乐的空形式按目标方言填写”。这是明显的语言/方言约束，但文档没有提供用户选择、opt-in 机制，且未说明该技能仅限特定地区或语言场景，因此构成自然语言层面的语言/locale 政策风险。
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This markdown file contains user-facing natural-language guidance exclusively in Chinese, and it does not provide an opt-in, alternative language, or explanation that the skill is intentionally region- or locale-specific. Under the policy, forcing a specific language without user choice is a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The text states that every video body should directly declare requirements in the project's prompt language, and the entire document is written as a Chinese production reference without offering a user language choice. This can amount to a locale/language policy constraint if the skill is used in contexts where users are not explicitly opting into Chinese-language prompt authoring.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 83% confidence
Finding: The document title and the entire instructional content are written in Chinese, which effectively imposes a specific language on reviewers. The file does not offer an alternative language option or explain that the skill is intentionally region- or locale-specific.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The file instructs authors to write prompts in Chinese and requires dialogue to use only one spoken language, specifically framed around Chinese usage. This is a natural-language locale constraint, and the document does not offer opt-in, alternatives, or an explicit policy justification for forcing that language choice.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: Line L009 sets video_prompt_language to zh-CN, and later lines reinforce Chinese-only prompt/dialogue requirements. This is a natural-language locale policy constraint with no visible opt-in, alternative language option, or explicit region-specific justification in the file.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: Lines L096-L102 require dialogue to be written in Chinese and say that non-Chinese dialogue must be specially marked, while also describing upstream constraints that allow only Chinese. This imposes a language policy on outputs without presenting it as optional or justified for a specific compliance context.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: This file contains a Chinese-titled contract and explicit language-specific rules, including requirements about quoted Chinese text and Chinese source documents. Under the policy, forcing a specific language or locale without user choice can be a natural-language policy violation unless the constraint is clearly justified as region- or domain-specific.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 86% confidence
Finding: Line L036 sets video_prompt_language to zh-CN in the example minimal structure, which can be read as a default locale requirement. While later text discusses other possible languages, this example still presents a fixed language choice without explicit user opt-in or a stated region-specific need.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The file instructs users to use '中文自然指令' and presents the dialect as Chinese-only behavior. This is a natural-language locale constraint, and the document does not offer a language choice or explicitly justify the restriction as a region- or model-mandated policy exception.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The validator explicitly accepts mode == "song" and requires non-empty lyrics, even though the skill metadata says the skill should not create lyrics. That mismatch weakens policy enforcement: downstream users or agents can submit lyric-bearing specs that the checker marks valid, enabling out-of-scope content generation and potentially policy bypass in workflows that rely on this validator as a gate.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.