Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MandarinClips MCP Server Find 3 HSK 3 movie clips using 还是…吧 for a lesson plan"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Official Model Context Protocol (MCP) server for MandarinClips — the linguistic cinema corpus designed for Chinese language educators (TCSL) and advanced learners.
Seamlessly connect Claude Desktop, Cursor, and AI agents directly to a database of 180,000+ authentic Chinese movie and television dialogue snippets, complete with native audio citations, HSK 3.0 grammar formulas, and real-world pragmatic subtexts.
Traditional textbooks teach sterile Chinese dialogues. Real native speech is packed with subtle cultural subtext, conversational hedging, and colloquial idioms.
This MCP server empowers LLMs to design authentic, cinema-backed Chinese lesson plans, exercises, and vocabulary guides:
Authentic Cinema Context: Real video turn sequences (previous, target, and succeeding lines) instead of robotic machine-generated sentences.
Pedagogical Indexing: Search by HSK 3.0 level (1 to 7-9), syntactic grammar patterns, or pragmatic social scenarios (e.g., polite refusals, workplace disputes).
Direct Markdown Citations: Every retrieved example automatically embeds an authentic scene link pointing directly to MandarinClips.
Related MCP server: video-material-db
No Node.js or installation required! In any MCP-compatible client or web agent (Claude Web, Smithery, Cursor Remote):
Server URL: https://www.mandarinclips.com/mcp
API Key Header: Authorization: Bearer <your_api_key> (or URL query: ?api_key=...)
Add this configuration to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"mandarinclips": {
"command": "npx",
"args": ["-y", "mandarinclips-mcp@1.1.0"],
"env": {
"MANDARINCLIPS_API_KEY": "mc_live_your_api_key_here"
}
}
}
}Add this configuration to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Run directly via npx with zero installation:
{
"mcpServers": {
"mandarinclips": {
"command": "npx",
"args": ["-y", "mandarinclips-mcp"]
}
}
}To get 8 free queries per day or unlimited Pro access, obtain your API key from your MandarinClips Account Dashboard:
{
"mcpServers": {
"mandarinclips": {
"command": "npx",
"args": ["-y", "mandarinclips-mcp"],
"env": {
"MANDARINCLIPS_API_KEY": "mc_live_your_api_key_here"
}
}
}
}Restart Claude Desktop, and you will see the MandarinClips tools ready in the toolbar!
Tool Name | Description | Parameters |
| Search 180,000+ real movie dialogues by Chinese characters, Pinyin, or English topics. |
|
| Retrieve structural sentence formulas and real dialogue examples for a grammar pattern. |
|
| Look up colloquial idioms, slang, and cultural nuances in cinematic contexts. |
|
| Retrieve official HSK 3.0 vocabulary lists with definitions and linked video counts. |
|
| Detailed dictionary profile of a Chinese character/word, showing linked cinema clips. |
|
"Design a 45-minute lesson plan for HSK 3 students on the '还是…吧' structure. Use MandarinClips to find 3 real movie dialogue examples."
"Give me 5 authentic video examples showing how native Chinese speakers politely refuse an invitation in daily life."
"Extract Level 2 HSK words related to food, provide pinyin, definitions, and link their video citations."
MandarinClips enforces fair-use daily quotas:
Anonymous (No API Key): 3 queries / day (per IP address).
Free Account: 8 queries / day. Get your free API key.
Pro Subscribers: Unlimited queries & high-speed exports via Patreon.
Educators & Professors: Apply for a complimentary 3-month Pro license via the Academic Access Program.
🌐 Web Platform: https://www.mandarinclips.com
🎓 HSK 3.0 Curriculum: Explore HSK Word Atlas
📐 Grammar Library: Browse Sentence Patterns
📌 Expressions & Idioms: Colloquial Phrase Directory
🎭 Pragmatic Scenes: Real-world Social Scenarios
📰 Official Blog: Insights on Chinese Pedagogy
MIT License © 2026 MandarinClips Team.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | Expression slug (e.g. 'hai-shi-suan-le-ba', 'bu-de-bu'). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | Grammar slug (e.g. 'hai-shi-ba' for '还是…吧', 'lian-dou' for '连…都…'). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| page | No | Page number for pagination. | |
| level | Yes | HSK 3.0 Level (1 to 7). Level 7 covers bands 7-9. | |
| limit | No | Number of vocabulary items per page. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | Word slug identifier (e.g. 'ai-6', 'zuo-tian'). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Max number of dialogue clips to return (1-20). | |
| query | Yes | Search keywords: Chinese characters (e.g. '算了'), Pinyin (e.g. 'suan le'), or English meaning/intent. | |
| hsk_level | No | Filter by HSK 3.0 Level (1 to 7). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.1.0get_expression_guideget_grammar_guideget_hsk_vocabularyget_word_detailsearch_cinema_clipsScored across 5 tools
Tool names suggest distinct resource types (clips, grammar, expressions, HSK vocab, word detail), but with zero descriptions an agent must guess boundaries. get_word_detail and get_hsk_vocabulary plausibly overlap on word-level lookups, and get_expression_guide vs get_grammar_guide could be confused.
All names are snake_case verb_noun, with four get_* tools and one search_* tool. The single search_ prefix is a sensible deviation for a query-style operation, so the pattern is largely predictable.
Five tools is a tight, well-scoped surface for a niche Mandarin-learning clip server. Nothing looks redundant enough to be padding, though the set is on the thin side.
Core content types (clips, grammar, expressions, HSK vocab, word detail) are covered, but there is no way to search expressions or vocabulary, no translation or example-sentence lookup, and no practice/quiz operation. Agents may hit dead ends for common language-learning tasks.
Search Chinese TV drama scenes with second-level timestamps by character, emotion, or scene type.
YouTube transcripts, search, channel/playlist listings and upload tracking for AI agents.
Chinese dictionary (124k entries), HSK levels, text segmentation and 195 grammar points.
Video analysis AI: transcripts, summaries, visual scenes/shots, clips, answers in natural language.
