superbuy-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SUPERBUY_CLAUDE_MODEL | No | Pick a model used by the Claude integration, for example `sonnet`. |
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 |
|---|---|
| statusC | Status van bridge en Firefox-extensie. |
| searchA | Zoek op Superbuy. platform: 'tb' (Taobao/Tmall; affiliate-zoekmachine, veel ruis), '1688' (groothandel, betere relevantie) of 'xianyu' (tweedehands). Chinese zoektermen werken het best (translate gaat dan uit). extra: extra/overschrijvende formuliervelden (sortering, prijsfilter e.d.; zie superbuy_captures(grep='crawler') voor wat de site zelf stuurt). Geeft compacte regels: id | prijs € (¥) | verkopen | winkel | Chinese titel | Google-titel. |
| detailsA | Productdetails van een item uit search() (id zoals 'tb:123'): alle varianten met eigen prijs (ontmaskert lokprijzen), verzending binnen China, voorraad, winkelscores, Superbuy-waarschuwingen. Telt zwaar mee voor Superbuy's rate-limit: alleen voor serieuze kandidaten gebruiken. |
| superbuy_fetchC | Ruwe fetch binnen de Superbuy-tab (alleen *.superbuy.com), voor API-verkenning. |
| superbuy_capturesB | API-verkeer dat de extensie passief zag (url-regex grep). Toont request en begin van response. |
| publishB | Publiceer een gecureerde selectie naar de resultatenpagina. items: [{id, title_nl, notes, verdict: 'top'|'goed'|'misschien'|'nee', score?: 0-10}] — id zoals uit search(); prijs/foto/links worden aangevuld uit de cache. request: de oorspronkelijke wens van de gebruiker. queries: gebruikte zoektermen. session_id: meegeven om een bestaande sessie bij te werken. |
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 6 tools
Search, details, and publish form a clear pipeline, and status is diagnostically distinct. The only mild overlap is between superbuy_fetch (active raw fetch) and superbuy_captures (passive traffic log), which could be confused by a hurried agent, but descriptions clarify the difference.
The two superbuy_-prefixed snake_case tools sit alongside bare nouns (status, search, details, publish), so there's no single predictable pattern. Names remain readable and inferable, but the mix of prefixed and unprefixed conventions is noticeable.
Six tools is well within the healthy range and each maps to a step in the research-to-publish workflow. The two diagnostic tools (status, captures) are somewhat auxiliary but justified given the bridge/extension architecture.
The surface covers the core lifecycle: search → details → publish, with diagnostics for the bridge and API exploration. Minor gaps exist (no session listing/cleanup, no order or cart interaction), but the stated curation workflow is fully supported.