Back to skill

Security audit

抖音文案一键智能提取

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a real cloud video transcription tool, but it needs Review because it can upload arbitrary local files and fetch arbitrary URLs despite narrower documentation.

Install only if you trust the publisher and service, and only run it on explicit video/media inputs you have reviewed. Avoid giving it sensitive local paths or untrusted URLs, especially in cloud or enterprise environments, until it adds media validation, URL/domain restrictions, upload-domain allowlisting, size limits, and clearer data-handling disclosures.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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 (4)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
scripts/utils/validator.js:11
Finding

Unrestricted Local File Selection Enables Sensitive File Upload

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/utils/validator.js:3
Finding

Server-Side Request Forgery Through Unrestricted URLs and Redirects

Content
View full analysis
this.__opts.maxRedirects) { const err = new Error("Too many redirects"); this.__setState(this.__states.FAILED); this.emit("error", err); return reject(err); } const redirectedURL = /^https?:\/\//.test(response.headers.location) ? response.headers.location : new URL(response.headers.location, url).href; this.emit("redirected", redirectedURL, url); return getRequest(redirectedURL, getReqOptions(redirectedURL)); } ``` The actual download path follows redirects in the same manner: ```js const redirectedURL = /^https?:\/\//.test(response.headers.location) ? response.headers.location : new URL(response.headers.location, this.url).href; this.__isRedirected = true; this.__initProtocol(redirectedURL); this.emit("redirected", redirectedURL, this.url); return this.__start(); ``` ### Technical Analysis Any syntactically valid HTTP or HTTPS URL is accepted. The code does not reject: - Loopback addresses - Private IPv4 ranges - IPv6 loopback or private ranges - Link-local addresses - Cloud metadata endpoints - Internal DNS names - Reserved or special-purpose address ranges - Hosts that resolve to internal addresses Redirects are followed up to the configured redir ...[truncated 2060 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/utils/upload.js:10
Finding

Server-Controlled Presigned URL Can Redirect Local File Uploads to Arbitrary HTTPS Hosts

Content
View full analysis
{ const fileStats = fs.statSync(filename); const totalSize = fileStats.size; let uploadedSize = 0; let settled = false; const fileStream = fs.createReadStream(filename); const uploadHeaders = Object.assign({}, headers); uploadHeaders["Content-Length"] = String(totalSize); if (!uploadHeaders["Content-Type"]) { uploadHeaders["Content-Type"] = "application/octet-stream"; } const options = { host: url.hostname, path: url.pathname + url.search, method: "PUT", headers: uploadHeaders, }; const req = https.request( { ...options, timeout: constants.REQUEST_TIMEOUT }, ``` The first error string above is translated into English; the surrounding executable behavior is reproduced directly. ### Technical Analysis The presigned URL and header object originate from the remote API response. The client only verifies that the URL uses the `https:` scheme. It does not validate: - The destination hostname - The destination port - Whether the hostname is an IP literal - Whether the destination belongs to the documented storage provider - Whether the destination resolves to an internal address - Whether all server-provided headers are safe to forward As a result, a compromised, malicious, or incorrectly configured API can instruct the client to upload the complete selected file to any HTTPS host. The server-provided header object is copied without an allowlist and forwarded to the sam ...[truncated 1650 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/utils/helper.js:27
Finding

Unbounded Remote Downloads Allow Disk and Network Resource Exhaustion

Content
View full analysis
{ const options = { retry: { maxRetries: constants.CREATE_MAX_ATTEMPTS, delay: constants.RETRY_INTERVAL, }, override: { skip: true, skipSmaller: true }, }; let progressLog = ""; const dl = new Downloader(url, path, options); ``` The downloader’s relevant defaults include no effective timeout: ```js this.__defaultOpts = { body: null, retry: false, method: "GET", headers: {}, fileName: "", timeout: -1, metadata: null, override: false, forceResume: false, removeOnStop: true, removeOnFail: true, maxRedirects: 10, progressThrottle: 1000, ``` The response is written directly to disk: ```js if (!this.__isResumed) { const _fileName = this.__getFileNameFromHeaders(response.headers); this.__filePath = this.__getFilePath(_fileName); this.__fileName = this.__filePath.split(path.sep).pop(); if (fs.existsSync(this.__filePath)) { const downloadedSize = this.__getFilesizeInBytes(this.__filePath); const totalSize = this.__total ? this.__total : 0; if ( typeof this.__opts.override === "object" && this.__opts.override.skip && (this.__opts.override.skipSmaller || downloadedSize >= totalSize) ) { this.emit("skip", { totalSize: this.__total, fileName: this.__fileName, filePath: this.__filePath, downloadedSize: downloadedSize, }); this.__setState(this.__states.SKIPPED); return resolve(true); } } this.__fileStream = fs.createWriteStream(this.__filePath, {}); } ``` ### Technical Analysis The download workflow ...[truncated 1970 chars]
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (23)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If the implementation only validates URLs and local file paths rather than performing the advertised transcription and content-generation tasks, the documentation overstates capabilities and may mask what the skill truly does. This is risky because orchestration systems may auto-select the skill for broad user intents, giving an untrusted or incomplete component access to user-supplied paths and URLs without delivering the claimed business function.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If the implementation only validates URLs and local file paths rather than performing the advertised transcription and content-generation tasks, the documentation overstates capabilities and may mask what the skill truly does. This is risky because orchestration systems may auto-select the skill for broad user intents, giving an untrusted or incomplete component access to user-supplied paths and URLs without delivering the claimed business function.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

If the implementation only validates URLs and local file paths rather than performing the advertised transcription and content-generation tasks, the documentation overstates capabilities and may mask what the skill truly does. This is risky because orchestration systems may auto-select the skill for broad user intents, giving an untrusted or incomplete component access to user-supplied paths and URLs without delivering the claimed business function.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If the implementation only validates URLs and local file paths rather than performing the advertised transcription and content-generation tasks, the documentation overstates capabilities and may mask what the skill truly does. This is risky because orchestration systems may auto-select the skill for broad user intents, giving an untrusted or incomplete component access to user-supplied paths and URLs without delivering the claimed business function.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

If the implementation only validates URLs and local file paths rather than performing the advertised transcription and content-generation tasks, the documentation overstates capabilities and may mask what the skill truly does. This is risky because orchestration systems may auto-select the skill for broad user intents, giving an untrusted or incomplete component access to user-supplied paths and URLs without delivering the claimed business function.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

If the implementation only validates URLs and local file paths rather than performing the advertised transcription and content-generation tasks, the documentation overstates capabilities and may mask what the skill truly does. This is risky because orchestration systems may auto-select the skill for broad user intents, giving an untrusted or incomplete component access to user-supplied paths and URLs without delivering the claimed business function.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · README.md (reported line 99)May include surrounding context.

md
4. 同时传入文件路径与任务ID,优先执行 `--id`,忽略 `--file`
5. 无自定义 prompt 时,默认完整转录视频全部文字

| 用户自然语言指令                                         | 自动生成命令                                                                                                |
| -------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| 视频提取 https://example.com/video.mp4 中的文字          | `node scripts/video2text/index.js --file "https://example.com/video.mp4"`                                   |
| 把本地 /path/to/your/video.mp4 改成小红书风格的文案      | `node scripts/video2text/index.js --file "/path/to/your/video.mp4" --prompt "改写成小红书风格的文案"`       |

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
74% confidence
Finding

The skill declares access to an environment variable (GUAIKEI_API_TOKEN) but does not define an explicit tool/permission scope such as allowed tools or permissions. That creates ambiguity about what execution capabilities the skill is intended to use and weakens reviewability, especially for a command-type skill that may run code and handle local files or network inputs.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger scope is very broad and covers many generic content-processing intents, making accidental or overbroad invocation likely. In an agent environment, overly broad routing can send unrelated user content, URLs, or local file paths to a third-party-integrated skill without sufficiently specific user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The instructions explicitly tell the AI to prioritize this skill for broad natural-language requests, which can override safer routing decisions and cause unnecessary third-party data exposure. This is particularly risky because the skill accepts URLs, local paths, and free-form prompts, so mistaken invocation can expand both data handling and execution scope.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · SKILL.md (reported line 121)May include surrounding context.

md
当用户用自然语言下达指令时,按以下映射生成命令,保证识别与执行一致:

| 用户自然语言指令                                     | 生成的命令                                                                                                  |
| ---------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| 提取 https://example.com/video.mp4 里的文字          | `node scripts/video2text/index.js --file "https://example.com/video.mp4"`                                   |
| 总结这个视频的核心观点 https://example.com/video.mp4 | `node scripts/video2text/index.js --file "https://example.com/video.mp4" --prompt "总结这个视频的核心观点"` |

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The package description advertises many loosely bounded use cases and input types, including local files, multiple third-party platform links, transcription, summarization, rewriting, translation, and content generation. Broad activation language increases the chance the skill is invoked in unintended contexts, causing over-collection of user content or execution on untrusted media/URLs without clear scope restrictions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This file uses Chinese-only natural-language comments, parameter descriptions, operation labels, and error messages throughout, with no indication that the skill is region-specific or that users can choose another language. That creates a language/locale policy issue because the skill implicitly forces a specific language experience without opt-in or justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This JavaScript file contains user-facing comments, error messages, and help text entirely in Chinese, including the top-level description and all CLI guidance. Under the stated policy, forcing a specific language without offering a user choice or documenting a justified locale constraint is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This file implements a fully generic HTTP/HTTPS downloader with redirects, retries, resume support, arbitrary headers, optional request body, and caller-controlled filenames. That capability is broader than the skill’s stated purpose of Douyin transcript/copy extraction and can be repurposed to fetch arbitrary remote content, increasing the risk of SSRF-style access to internal resources or unauthorized downloading if any user-controlled URL reaches this class.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This JavaScript file contains natural-language content in comments and runtime status/error messages exclusively in Chinese, such as the function comment and multiple inlineLog strings. Because the skill does not provide any user opt-in, fallback, or documented justification for a Chinese-only locale, it violates the language/locale policy criterion.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

Multiple user-facing error strings are hard-coded in Chinese, including request failures and token guidance. This enforces a specific language/locale without any opt-in, fallback, or justification, which matches the policy category for language/locale violations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file emits all user-facing warning and status messages in Chinese only, including the token misconfiguration notice and recovery instructions. This imposes a specific language on users without offering a language choice or documenting that the skill is intended only for a Chinese-speaking or region-specific audience.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The comment makes a strong privacy and retention promise that uploaded videos are not retained and are automatically deleted after transcription, but this file only performs an upload and contains no deletion, retention control, or verification logic. In a skill that handles user video content for transcription, this mismatch can mislead users and operators about how sensitive media is handled, creating privacy, compliance, and trust risk if files persist in storage longer than represented.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

This file contains natural-language comments and error strings exclusively in Chinese, including the upload privacy statement and runtime error text. Under the policy, forcing a specific language without user opt-in can be a locale-policy violation, and this file does not indicate that the skill is explicitly China-only or offers language selection.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The file presents all operational instructions, warnings, and examples only in Chinese. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale constraint is explicitly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

This JavaScript file contains its descriptive comment and all user-facing status/error messages exclusively in Chinese. Under the policy, forcing a specific language without user choice is a natural-language policy violation unless the locale restriction is explicitly documented and justified, which is not present here.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The banner text is presented only in Chinese, which constitutes a natural-language locale choice embedded in the code. There is no indication here that the skill is region-specific or that users can opt into a different language, so this may violate language/locale policy guidance.

Content

No source excerpt is available for this finding.