MCP server enabling HydraDB memory operations (store, recall, search) in AI coding tools
概览
MCP (Model Context Protocol) server for Hydra DB, the state-of-the-art agentic memory. Provides tools for storing, recalling, and managing memories with knowledge-graph enriched context. - — the npx @hydradb/mcp binary each client spawns. No server to operate; credentials live in the client's config. This is the default and everything below the Configuration section documents it. - — one hosted process behind a URL like https://mcp.hydradb.com that many clients point at, with nothing to install. See . Ids flow between these: hydradb_query, hydradb_list and hydradb_subgraph emit them; hydradb_inspect, hydradb_delete, hydradb_status, hydradb_subgraph and hydradb_relations accept them. Hydra DB also runs , queried in Cypher. This is a different product surface from the memory and knowledge above, and nothing crosses between them: hydradb_query cannot see graph data, and hydradb_graph_query cannot see memories.
README
以下为原文 README 节选，完整内容请查看原文。
Hydra DB — MCP Server
MCP (Model Context Protocol) server for Hydra DB, the state-of-the-art agentic memory. Provides tools for storing, recalling, and managing memories with knowledge-graph enriched context.
Run it two ways, same tools either way:
- Local (stdio) — the
npx @hydradb/mcpbinary each client spawns. No server to operate; credentials live in the client’s config. This is the default and everything below the Configuration section documents it. - Remote (HTTP) — one hosted process behind a URL like
https://mcp.hydradb.comthat many clients point at, with nothing to install. See Remote / hosted server.
Available Tools
| Tool | What it does |
|---|---|
hydradb_query |
Search memories and knowledge together, with knowledge-graph context |
hydradb_ingest |
Save a note, a document, or a conversation |
hydradb_list |
Enumerate what is stored — memories, knowledge sources, or both in one page |
hydradb_inspect |
Fetch one source’s full content by id |
hydradb_delete |
Remove one or more items by id, irreversibly |
hydradb_status |
Check whether an ingested source has finished indexing |
hydradb_subgraph |
Everything connected to one item — its thread, replies, parents, children, links |
hydradb_relations |
The knowledge-graph relations of one item or a whole collection, page by page |
hydradb_feedback |
Report whether a query’s results were useful, by its request_id |
Ids flow between these: hydradb_query, hydradb_list and hydradb_subgraph emit them;
hydradb_inspect, hydradb_delete, hydradb_status, hydradb_subgraph and hydradb_relations accept them.
Graph tools (Cypher)
Hydra DB also runs property graphs you model and own end to end, queried in Cypher. This is a different product surface from the memory and knowledge above, and nothing crosses between them: hydradb_query cannot see graph data, and hydradb_graph_query cannot see memories.
| Tool | What it does |
|---|---|
hydradb_graph_query |
Run Cypher — reads and writes |
hydradb_graph_collections |
List the graphs in a graph database |
hydradb_graph_admin |
Create a graph database; drop a collection or a database |
hydradb_graph_query is annotated destructiveHint, because it runs arbitrary Cypher and DELETE is as reachable through it as MATCH. There is deliberately no separate read-only Cypher tool and no read-only mode: both would mean classifying Cypher text client-side to decide what to refuse, which is a heuristic — a promise the server can keep and a client cannot. This server does not inspect your query at all; it sends it and reports what HydraDB says. To lock the graph surface down, withhold the tools (below) — that is a real guarantee.
// Everyone Alice knows within four hops安装
npx -y @hydradb/mcp@latest配置
{
"mcpServers": {
"hydradb": {
"command": "npx",
"args": ["-y", "@hydradb/mcp@latest"],
"env": {
"HYDRADB_API_KEY": "your-api-key",
"HYDRADB_DATABASE": "your-database"
}
}
}
}