io.github.cbetz/last-ehrsearch

Last EHR — Medplum

Verified · 12 days ago

FHIR chart tools for MCP: read-only by default, with elicitation-gated, human-approved writes.

Install

claude mcp add last-ehr -- npx -y @lastehr/[email protected]

Requires: MEDPLUM_ACCESS_TOKEN, MEDPLUM_CLIENT_ID, MEDPLUM_CLIENT_SECRET, MEDPLUM_BASE_URL

Our take

This is a read-only Medplum-oriented server for chart search and patient context inside MCP clients. It is for healthcare developers or internal clinical tooling that need safe retrieval rather than write access. The read-only design is a good restraint for a sensitive domain, but the project still looks small and should be vetted carefully for compliance and operational maturity.

reviewed by hand · 2026-07-12

Something wrong or dead here? Report it

Verification record

last verified
12 days ago
github stars
4
last commit
33 days ago
archived
no
license
Apache-2.0
latest version
0.3.1
in registry since
2026-07-26

github.com/cbetz/last-ehr

mcp-server-searchsearch

mcp-server-search

Verified · 27 days ago

MCP server for web search operations

hand-reviewed39k starschecked 27 days ago

repomixsearch

Repomix

Verified · 18 days ago

Pack local or remote codebases into AI-friendly files that LLMs and coding agents can read or search

hand-reviewed28k stars67k dl/wkchecked 18 days ago

screenpipe-mcpsearch

screenpipe

Verified · 20 days ago

Search your local screen recordings, audio transcripts, and computer activity from screenpipe.

hand-reviewed22k stars2.8k dl/wkchecked 20 days ago

sui-kapasearch

Sui Knowledge Base (kapa.ai)

Verified · 14 days ago

Semantic search across Sui documentation and knowledge sources, powered by kapa.ai

hand-reviewed7.8k starschecked 14 days ago

firecrawl-mcp-serversearch

Firecrawl MCP Server

Verified · 34 days ago

MCP server for Firecrawl — web search, scraping, and biomedical/arXiv paper search.

hand-reviewed7.4k starschecked 34 days ago

robinsearch

Robin: AI-Powered Dark Web OSINT

Verified · 14 days ago

Dark web OSINT over Tor: search onion engines, scrape pages, report with your own model.

hand-reviewed7.3k starschecked 14 days ago