What it does
The server exposes Kling AI video generation as MCP tools. It supports text-to-video, image-to-video using reference start and end frames, extending an existing video, and transferring motion from a reference video to a character image. It offers camera movement control across nine Kling models, and tools to check the status and result of a single generation task or several at once. Helper tools list the available models and API actions.
It suits someone who wants to create or extend AI video directly from an MCP-compatible assistant or editor, such as Claude or VS Code, without leaving the chat. Task tracking is useful when generations take time and the results need to be retrieved later.
Setup offers three routes: a hosted endpoint with OAuth sign-in, a hosted endpoint with an AceDataCloud API token sent as a Bearer header, or a local stdio process that receives the token through an environment variable. After connecting, the reference tool can confirm the tools are visible without calling a generation API. Checking authenticated access can be done by looking up a task the application already owns. Each generation is a separate operation, and the publisher advises reviewing current pricing before running one.
Limitations and requirements
Requires an AceDataCloud account and an application key or OAuth sign-in with appropriate service access. The hosted service is metered. The reference tool does not validate the key, account balance or current price, and generation requests can create charges. Keys can be restricted by expiration, usage limit and allowed APIs, and access errors may need those settings checked. Dynamic Client Registration is not an API key. Do not configure OAuth and a fixed Authorization header for the same server. The local stdio route still calls the AceDataCloud API and needs the token passed to the process.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 15 Nov 2026 · its page on the source
io.github.AceDataCloud/mcp-kling
MCP server for Kling AI video generation
v2026.11.15.0As shown on registry.modelcontextprotocol.io when we last read it