tourwright
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| TOURWRIGHT_API_URL | No | URL of a remote Tourwright API to point the MCP server at. | |
| TOURWRIGHT_API_TOKEN | No | API token for the remote Tourwright API. | |
| TOURWRIGHT_MCP_ALLOW_FILES | No | Set to true to enable the tourwright_render_file MCP tool. | false |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| tourwright_guideA | The script format (browser walkthroughs and narrated slides), every action, voices and examples. Read it before writing a script. |
| tourwright_voicesA | Text-to-speech voices available on the server and which engines work. Use a voice whose language matches the script. |
| tourwright_validateA | Check the structure of a script (fields, actions, variables) without opening any website: fast, offline. Returns { ok } or the list of errors to fix. To know whether its targets still exist on the live site, use tourwright_check. |
| tourwright_renderA | Queue a script for rendering. With wait=true (default) it waits until the video is ready and returns its links (and local paths), duration, chapters and captions; otherwise it returns the job id to poll with tourwright_job. |
| tourwright_jobB | Status, progress and (when done) the result of a render job. |
| tourwright_jobsA | The render jobs on this server, newest first: id, status, progress and, for finished ones, the links to their videos. Use it to find a video rendered earlier or a job id to delete. |
| tourwright_delete_jobA | Delete a finished (done or failed) job and its files: the video, subtitles, poster and manifest. A job that is queued or rendering cannot be deleted. |
| tourwright_checkA | Run a script fast, without voice or video, against the real site (validate only checks its structure). Returns every step that failed (the target it looked for, the page, the targets that page offers) and, when it can tell, the new target for it. For a YAML script given as text it also returns |
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 8 tools
Each tool maps to a distinct resource or action, and the descriptions explicitly contrast the two closest pairs (validate vs check: offline structure vs live targets; render vs check: video output vs failure report). The only friction is the singular/plural pair tourwright_job vs tourwright_jobs, which relies on the reader noticing one letter.
All names share the tourwright_ prefix and snake_case, with verb-style tools (validate, check, render, delete_job) alongside plural-noun list/query tools (jobs, job, voices, guide). This is a readable REST-like pattern, but the mix of explicit verbs and bare nouns (no list_ prefix) is not a single uniform convention.
Eight tools is well within the sweet spot and each one earns its place: two-step validation, dry-run checking, rendering, job inspection, cleanup, documentation, and voice discovery. Nothing looks like filler or a duplicated capability.
The authoring-to-render lifecycle is largely covered (guide, voices, validate, check with auto-fix, render, list/get/delete jobs). However, delete_job explicitly refuses queued or rendering jobs and there is no cancel/stop tool, so a running render is a dead end the agent cannot recover from; there is also no re-render or script retrieval operation.