SceneWeaver MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SCENEWEAVER_HOME | No | Set in the MCP process environment to choose where state, media and results are stored; doctor reports the active location. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_capabilitiesA | Report installed voices, provider options, limits and privacy boundaries. |
| search_mediaC | Find licensed candidates; consult source pages and notices before selecting them. |
| import_mediaB | Import only a filename staged by the operator in the managed inbox, with rights metadata. |
| create_storyboardA | Plan explicit scenes: text, query, optional asset_id and optional duration_seconds. The assistant writes the scenes; no model API is called. Set every scene duration or let the planner divide the total by word count. Branding supports title, footer, color and a managed logo_asset_id. Shorten text if narration cannot fit naturally; speech is never silently cut. |
| get_storyboardC | Read a saved plan and its scene-to-media choices. |
| select_scene_assetA | Replace one scene's media choice without regenerating its script or other choices. |
| preview_storyboardB | Return a contact sheet inline and its scene timeline for semantic review. |
| generate_videoA | Start an exact-duration render of the selected storyboard; poll get_video_status. |
| preview_voiceB | Create a short local WAV preview before choosing a narrator. No text leaves the computer. |
| get_video_statusD | Read progress and a bounded actionable failure message. |
| get_video_resultsB | Retrieve the MP4, SHA-256, credits, quality report and measured duration. |
| cancel_videoB | Cancel this job and terminate its render process tree. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 12 tools
Most tools target clearly distinct actions in the media-to-video pipeline, such as search vs. import and generate vs. status/results/cancel. There is some potential overlap between get_storyboard and preview_storyboard, and between media search/import/select, but the descriptions make the boundaries reasonably clear.
All tool names use consistent snake_case with a predictable verb_noun structure. Related lifecycle operations are grouped coherently, such as get_video_status, get_video_results, and cancel_video.
The 12 tools are well-scoped for a storyboard-to-video workflow, covering media sourcing, planning, preview, generation, monitoring, results, cancellation, and capabilities. Each tool appears to earn its place without excessive surface area.
The core workflow is covered end to end, including media intake, storyboard creation and preview, asset selection, rendering, status polling, results, and cancellation. Minor gaps exist around editing or deleting storyboards beyond media selection, and there is no listing of prior jobs or imported media.