Backenly
io.github.backenly/backenly
Postgres backend your coding agent builds over MCP; destructive actions wait for a person.
Remote endpoint
https://backenly.com/api/mcpOLACHEAP · MCP WORKSPACE / DATA
Data MCP servers help query databases, warehouses, and analytics systems, but a conversational answer should be traceable to actual tables, permitted fields, and a known time window.
Results come from the current synchronized Registry snapshot, not popularity or activity.
io.github.backenly/backenly
Postgres backend your coding agent builds over MCP; destructive actions wait for a person.
https://backenly.com/api/mcpio.github.bartholomej/node-csfd-api
Movies, TV series, creators and user ratings from ČSFD.cz, the Czech-Slovak film database
io.github.benchoi93/refcheck
Verify academic citations against real databases. Catches hallucinated references, exports BibTeX.
io.github.benseverndev-oss/goldenflow
Standardize, reshape, and normalize messy data — CSV, Excel, Parquet, S3, databases.
https://goldenflow-mcp-production.up.railway.app/mcp/io.github.bersimao/b1-mcp-server
MCP server for SAP Business One database (HANA / MS SQL) and Service Layer operations
io.github.bguillow-rgb/percolate
Search the Percolate specialty coffee database: 1,100+ coffees with brew recipes and pairings.
https://mcp.percolateapp.com/io.github.bguillow-rgb/perfume-picks
Search the Perfume Picks fragrance database: 10,000+ scents with notes, dupes, and wear data.
https://mcp.perfumepicks.app/io.github.bguillow-rgb/pour-picks
Search the Pour Picks bourbon & whiskey database: 4,700+ bottles with tasting profiles and prices.
https://mcp.pourpicks.app/io.github.billy12151/memory-arbiter-mcp
Local SQLite MCP memory: evidence-based recall, advisory conflict notices, authorized governance.
io.github.bluwork/mongo-scout-mcp
Scout your MongoDB databases with AI - safety features, live monitoring, and data quality
io.github.bluwork/postgres-scout-mcp
Scout your PostgreSQL databases with AI - safety features, monitoring, and data quality
io.github.bolnet/memwright
Embedded memory for AI agents with SQLite, pgvector, and Neo4j graph search.
io.github.boringSQL/dryrun
PostgreSQL schema intelligence from a committed snapshot. No connection, no credentials.
io.github.bpamiri/u2-mcp
Connect AI assistants to Rocket Universe/UniData MultiValue databases via MCP.
io.github.bricksentry/bricksentry
Safety check for AI agent and pipeline writes to Databricks tables, before they reach production.
io.github.browtastic/freellms-gateway
FreeLLMs MCP gateway: filesystem, database, shell, fetch, auth and devtools.
https://mcp.freellms.net/sseio.github.bruno-portfolio/hypokrates
Pharmacovigilance hypothesis generation from 15 public drug safety databases
io.github.bvenkata/legacy2mcp
Turn a legacy SOAP/WSDL, database, or REST API into a safe, typed, audit-logged MCP server
io.github.bytebase/dbhub
Token-efficient MCP server for PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, ClickHouse, SQLite
io.github.caiderek/logcarver-mcp
Read-only MCP server that finds deleted SQL Server rows from the transaction log (fn_dblog).
io.github.caiderek/logcarverguard
Guardrail for AI agents on SQL Server: statement blocking, optional read-only mode and an audit log.
io.github.calasdan-sketch/monomem
Local shared memory for AI coding agents: SQLite notes with writer provenance, undo, search.
io.github.ccedacero/nyc-property-intel
MCP server giving Claude AI access to 22+ NYC public-record databases for real estate due diligence
https://nyc-property-intel-production.up.railway.app/mcpio.github.chenshiyilisha/multidb-mcp
MCP server for Dameng (DM8), SQL Server, and MySQL — read-only by default, granular write perms
Synchronized data; no invented downloads, reviews or ratings.
OLACHEAP · PRACTICAL GUIDES
Start with a small read-only sample and explicit table names. Request the query plan or statement when supported, check row counts and time zones, and compare aggregates against an authoritative report.
Check database roles, row-level access, connection transport, query limits, result truncation, and whether the tool can modify schema. Avoid granting production administrator credentials to a server used for exploratory analysis.
A model can generate a plausible but incorrect SQL query or misinterpret null values. Log queries with sensitive values redacted, require approval for write statements, and prevent confidential rows from being sent to an unintended provider.
Use their exact Registry IDs to compare published connection information before granting permissions.