BlueVetaUpright MCP server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BLUEVETA_UPRIGHT_API_KEY | No | Your BlueVetaUpright API key (BYOK), e.g. bvu_live_.... Set as an environment variable in your MCP client's server config. A call can alternatively override the key per-invocation with the tool's apiKey argument. |
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 |
|---|---|
| detect_rotationA | Checks whether a product photo is upright and, if not, exactly how many degrees to rotate it (clockwise). Returns a confidence score and an honest flag when the answer is uncertain. Bills the caller's own BlueVetaUpright account per call, same as the REST API -- requires an API key from https://blueveta.com/upright, set as BLUEVETA_UPRIGHT_API_KEY or passed as apiKey. |
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 1 tool
There is only one tool, so there is no possibility of misselection or overlap. The tool's stated purpose (upright/rotation-angle detection with confidence and uncertainty flagging) is unambiguous.
detect_rotation follows a clear verb_noun snake_case convention. With a single tool there is no risk of mixed conventions, so the set is trivially consistent.
One tool is thin for a server surface, but the declared domain is genuinely narrow (rotation detection only), so it is borderline rather than a clear mismatch. It lands at the 'feels thin' level rather than being grossly under- or over-scoped.
The detection operation itself is well covered (angle, confidence, uncertainty flag, auth/billing notes), but there are notable gaps: no batch/multi-image handling, no tool to apply the rotation or return a corrected image, and no usage or quota check. Agents can work around these but will hit a dead end for any follow-on step.