Search Vascue's public healthcare-ops, insurance-claims and booking docs. Public content only.
io.vascue/public-knowledge-search MCP Server
This Model Context Protocol (MCP) server searches Vascue’s public healthcare-ops, insurance-claims, and booking documentation. It is limited to public content and provides a single tool (toolCount: 1) for knowledge search across those doc areas.
🛠️ Key Features
Searches public healthcare-ops, insurance-claims, and booking docs
Exposes 1 MCP tool
Uses the Model Context Protocol (MCP) framework
🚀 Use Cases
Finding public guidance related to healthcare operations
Searching for information in insurance-claims documentation
Locating booking-related public documentation
⚡ Developer Benefits
Centralizes search over Vascue’s public knowledge sources
Integrates with MCP (Model Context Protocol) workflows
Topic coverage includes clinic, healthcare, insurance-claims, and booking-related docs
⚠️ Limitations
Searches only Vascue public content (no private or restricted material implied)
Hybrid keyword + semantic search over the public pages of vascue.io: healthcare-operations guides, Vascue Front Desk for clinics, provider-side insurance-claims automation, practice-management integrations (two-way sync with Cliniko and Nookal; Jane, Mindbody, PracticeHub, HexaHealth, Halaxy, Clinicea, ClinicOffice, Semble, Plato Medical and JERA in progress), security and compliance pages, case studies, pricing and blog posts.
Use it to answer questions about what Vascue offers, how its products work and what it has published. One topic per call; cite the returned page URL for every excerpt you use.
Returns a JSON text payload: result.chunks[] ordered by relevance, each with text (a Markdown excerpt), score (0-1, higher is better) and item.metadata.url (the canonical https://www.vascue.io/... page). Off-topic queries still return the nearest chunks, so treat scores below about 0.5 as weak evidence and say so when the site does not cover a topic.
Read-only and idempotent; no authentication; 60 requests per minute per client. Public content only: never send patient information, claim documents, clinic credentials or booking requests. It cannot book appointments or look up clinic data.
Parameters2
query
string
required
What to look for, as a natural-language question or keywords, e.g. "how does claims pre-authorisation work" or "Cliniko integration". 3-15 words works best; one topic per call.
ai_search_options
object
optional
Optional retrieval tuning. Omit for the defaults: hybrid search, 8 results, score threshold 0.35.
Raw schema
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "What to look for, as a natural-language question or keywords, e.g. \"how does claims pre-authorisation work\" or \"Cliniko integration\". 3-15 words works best; one topic per call."
},
"ai_search_options": {
"type": "object",
"description": "Optional retrieval tuning. Omit for the defaults: hybrid search, 8 results, score threshold 0.35.",
"properties": {
"retrieval": {
"type": "object",
"properties": {
"retrieval_type": {
"type": "string",
"enum": [
"hybrid",
"vector",
"keyword"
],
"description": "hybrid (default) fuses keyword and vector ranking; keyword for exact phrases or product names; vector for paraphrased questions."
},
"max_num_results": {
"type": "integer",
"minimum": 1,
"maximum": 50,
"description": "Maximum chunks to return (default 8)."
},
"match_threshold": {
"type": "number",
"minimum": 0,
"maximum": 1,
"description": "Minimum relevance score for a chunk to be included (default 0.35); raise for precision, lower for recall."
},
"context_expansion": {
"type": "integer",
"minimum": 0,
"maximum": 3,
"description": "Neighbouring chunks from the same page to include around each hit (default 0)."
}
}
}
}
}
},
"required": [
"query"
]
}
A Model Context Protocol server that searches Vascue's public documentation: healthcare operations, Vascue Front Desk for clinics, provider-side insurance claims automation, practice-management integrations (two-way sync with Cliniko and Nookal; Jane, Mindbody, PracticeHub, HexaHealth, Halaxy, Clinicea, ClinicOffice, Semble, Plato Medical and JERA in progress), security, case studies and pricing.
It comes in two equivalent forms:
Hosted (always current):https://www.vascue.io/mcp/search - streamable HTTP, no authentication.
Self-contained (this repo):python server.py - local BM25 search over a bundled snapshot of the public pages (content/, refreshed per release with scripts/fetch_content.py). No network calls at runtime, so it also works offline and is what directory-built releases run.
Endpoint:https://www.vascue.io/mcp/search (streamable HTTP, no authentication)
Public content only. This server indexes public product and educational pages. Never send patient information, claim documents, clinic credentials or booking requests to it. Agent-based clinic booking is a separate research pilot, not a public API.
Connect
Any MCP client that speaks streamable HTTP can connect to the endpoint directly.
Claude Code
bash
claude mcp add --transport http vascue-search https://www.vascue.io/mcp/search
Cursor / Claude Desktop / other stdio-only clients (via mcp-remote)
Hybrid (keyword + vector) search over Vascue's public pages. Returns matching excerpts with their canonical https://www.vascue.io/... URLs so answers can cite the source.
Input
Type
Notes
query
string (required)
Natural-language question or keywords, e.g. "how does Vascue handle insurance claim pre-authorisation".
ai_search_options.retrieval.retrieval_type
"hybrid" | "vector" | "keyword"
Default hybrid.
ai_search_options.retrieval.max_num_results
integer 1–50
Default 8.
ai_search_options.retrieval.match_threshold
number 0–1
Default 0.35.
ai_search_options.retrieval.context_expansion
integer 0–3
Neighbouring chunks to include.
Query rewriting and reranking are disabled server-side; the server returns source chunks only and never a generated answer, so nothing is presented as a Vascue statement without a citation. Rate limit: 60 requests per minute per client.
Example call:
json
{"name":"search","arguments":{"query":"Cliniko integration for Vascue Front Desk"}}
The endpoint is backed by a Cloudflare AI Search instance over the approved public Markdown export of vascue.io (the service descriptor at https://www.vascue.io/.well-known/ai-search.json states what is and is not indexed).
Local server (server.py)
BM25 keyword search over the bundled content/ snapshot. No network calls at runtime. Returns {"chunks": [...]}, each chunk with url, title, score (0–1, relative to the best match) and text.
Input
Type
Notes
query
string (required)
Natural-language question or keywords; 3–15 words, one topic per call.
CI runs the same build and smoke test on every push and weekly, so the badge above doubles as an endpoint health indicator.
Directory build specs
Directories that build the server from source (e.g. Glama) run the self-contained form. Generated build images vary (uv-managed Python without pip, or a PEP 668 externally-managed system Python), so use an explicit venv: