Click on "Install 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., "@Heygen MCP Server generate a video with avatar Mark saying hello in a friendly tone"
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.
The HeyGen MCP server enables any MCP Client like Claude Desktop or Agents to use the HeyGen API to generate avatars and videos.
Note: This project is in early development. While we welcome community feedback and contributions, please be aware that official support is limited.
Python 3.10 or higher
A Heygen API key (get one from Heygen). Includes 10 Free Credits per Month
uv is a fast Python package installer and resolver that we recommend for installing this package.
macOS or Linux:
# Install with the official installer script
curl -LsSf https://astral.sh/uv/install.sh | sh
# Or via Homebrew (macOS)
brew install uvWindows:
# Install with the official installer script in PowerShell
irm https://astral.sh/uv/install.ps1 | iex
# Or via Scoop
scoop install uvFor other installation methods, see the uv documentation.
Related MCP server: heygen-mcp
Get your API key from HeyGen.
Install uv package manager (see Installing uv section above).
Go to Claude > Settings > Developer > Edit Config > claude_desktop_config.json to include the following:
{
"mcpServers": {
"HeyGen": {
"command": "uvx",
"args": ["heygen-mcp"],
"env": {
"HEYGEN_API_KEY": "<insert-your-api-key-here>"
}
}
}
}If you're using Windows, you'll need to enable "Developer Mode" in Claude Desktop to use the MCP server. Click "Help" in the hamburger menu at the top left and select "Enable Developer Mode".
The server provides the following tools to Claude:
get_remaining_credits: Retrieves the remaining credits in your Heygen account.
get_voices: Retrieves a list of available voices from the Heygen API (limited to first 100 voices).
get_avatar_groups: Retrieves a list of Heygen avatar groups.
get_avatars_in_avatar_group: Retrieves a list of avatars in a specific Heygen avatar group.
generate_avatar_video: Generates a new avatar video with the specified avatar, text, and voice.
get_avatar_video_status: Retrieves the status of a video generated via the Heygen API.
To run the server locally with the MCP Inspector for testing and debugging:
uv --with "mcp[cli]" dev heygen_mcp/server.pyThis will start the server in development mode and allow you to use the MCP Inspector to test the available tools and functionality.
Tests
CICD
Photo Avatar APIs Support
SSE And Remote MCP Server with OAuth Flow
Translation API Support
Template API Support
Interactive Avatar API Support
Contributions are welcome! Please feel free to submit a Pull Request.
This project is licensed under the MIT License - see the LICENSE file for details.
Generates a new avatar video via the HeyGen API.
| Name | Required | Description | Default |
|---|---|---|---|
| avatar_id | Yes | ||
| input_text | Yes | ||
| voice_id | Yes | ||
| title | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It merely states it 'generates' a video, without mentioning if the operation is synchronous or asynchronous, potential side effects, rate limits, or what happens with errors. This is insufficient for an AI agent to understand the tool's behavior.
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?
The description is extremely short (one sentence), but conciseness is not valuable when it omits critical information. It fails to convey key details about parameters, usage, or behavior.
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?
Given the tool has 4 parameters (3 required), no output schema, and no annotations, the description is completely inadequate. It provides no context about the return value, error handling, or how the video generation process works.
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?
The input schema has 0% description coverage, meaning none of the 4 parameters are described. The description adds no meaning beyond the schema—it only mentions the API name. The agent has no indication of what each parameter represents or its expected format.
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?
The description clearly states the action ('Generates') and the resource ('new avatar video') with the specific API (HeyGen). However, it does not differentiate from the sibling tool 'generate_avatar_video_with_multiple_scenes' beyond the name, leaving ambiguity about whether this tool supports only single scenes.
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?
No guidance is provided on when to use this tool versus alternatives like 'generate_avatar_video_with_multiple_scenes'. There is no mention of prerequisites, typical use cases, or required authentication context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generates a new avatar video via the HeyGen API. with multiple scenes.
| Name | Required | Description | Default |
|---|---|---|---|
| title | Yes | ||
| video_inputs | Yes | ||
| dimension | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Without annotations, the description must disclose behavioral traits. It only states it generates a video via HeyGen API, but omits details like asynchronicity (status check via sibling), credentials, rate limits, or side effects.
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?
The description is too short (a single incomplete sentence) to be an effective MCP description. While concise, it lacks necessary detail, crossing into under-specification.
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?
Given the tool's complexity (multiple scenes, nested input objects, no output schema), the description is severely incomplete. It does not explain how to construct scenes or what the response contains.
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?
Schema coverage is 0% and the description adds no parameter information. It fails to explain 'title', 'video_inputs', or 'dimension', despite the nested complexity.
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?
The description clearly states the tool generates an avatar video with multiple scenes, distinguishing it from the sibling 'generate_avatar_video' which likely handles single scenes. The verb 'generates' and object 'avatar video' are precise.
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?
No guidance is provided on when to use this tool versus alternatives. For example, it does not compare with generate_avatar_video or explain when multiple scenes are needed.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieves a list of HeyGen avatar groups. By default, only private avatar groups are returned, unless include_public is set to true. Avatar groups are collections of avatars, avatar group ids cannot be used to generate videos.
| Name | Required | Description | Default |
|---|---|---|---|
| include_public | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so description carries full burden. It discloses default behavior and a limitation, but does not mention if the operation is read-only, idempotent, or any side effects. Lacks safety and performance cues.
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?
Two sentences, no unnecessary words. Each sentence provides distinct information: purpose and a behavioral detail. Efficiently structured.
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?
Adequate for a simple tool with one parameter. Explains parameter and a constraint, but does not describe return structure, error conditions, or authentication requirements. With no output schema, more detail on the expected response would improve completeness.
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?
With 0% schema description coverage, the description compensates well by explaining the include_public parameter's meaning and default behavior. Adds value beyond the schema's type and name.
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?
Description clearly states it retrieves a list of avatar groups, specifying the default behavior (private only) and the effect of include_public. Distinguishes itself from sibling tools like get_avatars_in_avatar_group by stating avatar group ids cannot be used to generate videos.
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?
Provides some context by noting that avatar group ids cannot be used to generate videos, but does not explicitly contrast with sibling tools or specify when to use this tool over others.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieves a list of avatars in a specific HeyGen avatar group.
| Name | Required | Description | Default |
|---|---|---|---|
| group_id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It only states the basic retrieval action, omitting any details about side effects, required permissions, rate limits, or whether the tool is read-only. This lack of transparency is insufficient for safe invocation.
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?
The description is a single, concise sentence that immediately conveys the tool's core purpose. It is appropriately front-loaded and contains no extraneous information, though it could benefit from slightly more detail without becoming verbose.
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?
The description lacks essential context such as what an avatar group is, how to acquire a group_id, what the returned list contains, or error handling. For a tool with no output schema and no annotations, this leaves the agent underspecified about usage expectations.
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?
The schema has 0% description coverage for its single parameter 'group_id'. The tool description does not add any meaning beyond the parameter name, failing to explain what a group_id is, how to obtain it, or its expected format. This does not compensate for the missing schema descriptions.
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?
The description clearly states the verb 'retrieves' and the resource 'avatars in a specific HeyGen avatar group', which precisely indicates the tool's function. It distinguishes itself from sibling tools such as 'get_avatar_groups' (which retrieves groups, not avatars) and the generation tools.
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?
The description provides no guidance on when to use this tool versus alternatives. There is no mention of prerequisites, use cases, or scenarios where other tools would be more appropriate, leaving the agent without decision support.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieves the status of a video generated via the HeyGen API. Video status make take several minutes to hours depending on length of video and queue time. If video is not yet complete, status be viewed later by user via https://app.heygen.com/home
| Name | Required | Description | Default |
|---|---|---|---|
| video_id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided; description adds behavioral context about potential delays and alternative viewing via URL. Lacks details on error states, input validation, or polling behavior.
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?
Two sentences with no fluff; first sentence defines purpose, second adds useful context. Every sentence earns its place.
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?
Adequate for a simple tool but incomplete: no mention of return format (status values, possible errors) or additional details about the URL. More context would reduce ambiguity.
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?
Only one parameter (video_id) with 0% schema description coverage. Description does not explain how to obtain video_id or its expected format, relying on obviousness.
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?
The description clearly states the tool retrieves the status of a video generated via the HeyGen API, using a specific verb and resource. It distinguishes from sibling tools like generate_avatar_video and get_avatar_groups.
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?
Implicitly indicates it is used after video generation, with context on time delays and a fallback URL. No explicit exclusions or alternative guidance, but context is clear enough.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieves the remaining credits in heygen account.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must convey behavior. It indicates a read-only retrieval, but lacks details on caching, rate limits, or authentication requirements. For a simple zero-parameter tool, this is adequate but not comprehensive.
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?
A single, clear sentence conveys the entire purpose without any unnecessary words or repetition.
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?
For a simple credit retrieval tool, the description provides the core purpose. However, without an output schema, it fails to specify the return format (e.g., number, string), which would help the agent interpret results.
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?
Schema coverage is 100% (no parameters), so baseline is 3. The description adds no parameter information since none exist.
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?
The description clearly states the action ('Retrieves') and the resource ('remaining credits in heygen account'). It is distinct from sibling tools which deal with avatar generation and voice management.
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?
No explicit guidance on when to use this tool versus alternatives, but the purpose is straightforward and siblings are clearly different. The description implies usage for checking credits but does not specify conditions like before video generation.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieves a list of available voices from the HeyGen API. Results truncated to first 100 voices. Private voices generally will returned 1st.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so the description carries the full burden. It discloses two key behaviors: result truncation to first 100 voices and ordering (private voices first). This adds value beyond a simple 'list voices' statement. However, it does not mention rate limits, authentication, or error handling.
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?
Two short sentences, front-loaded with the main action. Every sentence adds value. No redundant or verbose phrasing.
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?
The tool is simple (0 params, no output schema). The description provides truncation and ordering, but does not explain the return format or fields of each voice object. Without an output schema, this information would help agents parse results. Missing this detail reduces completeness.
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?
The schema has no parameters (0 params, 100% coverage). Baseline for 0 params is 4. The description does not need to add parameter info. It neither adds nor detracts from what the schema already conveys.
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?
The description clearly states it retrieves a list of available voices from the HeyGen API. The verb 'retrieves' and resource 'list of available voices' are specific. It distinguishes from sibling tools (avatar generation, credits) which have different purposes.
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?
No explicit guidance on when to use this tool versus alternatives. The sibling tools are sufficiently different that confusion is unlikely, but the description does not clarify usage context, prerequisites, or when not to use it.
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.
v0.0.3generate_avatar_videogenerate_avatar_video_with_multiple_scenesget_avatar_groupsget_avatar_video_statusget_avatars_in_avatar_groupget_remaining_creditsget_voicesScored across 7 tools
The two generate_avatar_video tools have overlapping descriptions and could confuse agents about which to use for single vs. multi-scene videos. Other tools are well-distinguished.
All tool names follow a consistent verb_noun pattern with snake_case: generate_ for creation and get_ for retrieval, making the surface predictable.
With 7 tools covering video generation, status, avatars, voices, and credits, the count is well-scoped for a focused HeyGen integration.
Core workflows like video generation, status checking, and resource listing are covered. Missing are tools to list or manage generated videos, but the surface is largely complete.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
AI voice generation: text-to-speech and voice cloning from any MCP client.
MCP server for Google Veo AI video generation
MCP server for Hailuo (MiniMax) AI video generation
MCP server for Kling AI video generation