Astra DB
io.github.ericharePublished by GitHub account ericharev2.0.0Updated Oct 24, 2026
Astra DB tools for agents: schema, find, vector search, guarded writes, and interactive views.
Overview
Community Astra DB MCP server that lets an assistant inspect schemas, query and vector-search collections, and perform guarded writes.
- What you can do
- Check connection status and list databases
- Explore collections and tables, view database overviews and vectorize providers
- Query with find, vector search, count and distinct values
- Look up bundled Data API code examples
- Insert, update, delete, create collections, tables and indexes, and drop objects
- When to use it
- Useful when an assistant needs to work with an Astra DB database: exploring its schema, running reads and vector or hybrid searches, drafting Data API code, or making changes with confirmation. It also offers a one-command setup for several coding agents.
- Requirements
- Node.js 20+ for the npm package, or the hosted remote endpoint with OAuth. An Astra DB application token (ASTRA_DB_APPLICATION_TOKEN) and usually the Data API endpoint (ASTRA_DB_API_ENDPOINT); keyspace is optional. Credentials can come from the environment, a project .env, or the Astra CLI config.
Installation
In SourceWeft
- Open Astra DB in the dashboard and add it to a workspace.
- Connect the required credentials. They are stored privately and never shown on public pages.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.github.erichare%2Fastra-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.
{
"mcpServers": {
"astra-db": {
"type": "http",
"url": "https://astra-widgets-mcp.vercel.app/mcp"
}
}
}README
JEStats Astra DB Plugin · @erichare/astra-mcp
Unofficial, community-maintained Astra DB integration by JEStats, not affiliated with or endorsed by DataStax or IBM. An MCP server with live data tools, guarded writes, vector and hybrid search, interactive MCP Apps views, and about 1,660 bundled Data API examples. It also includes a one-command installer for Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop, and IBM Bob.
init detects your agents and configures each one. Its login step reuses an existing credential source or asks for an Astra application token with hidden input, lets you pick a database, and writes connection settings to a git-ignored .env. There's no restart: the server re-reads credentials on every call. Requires Node.js 20+.
Already configured the Astra CLI? Run login --profile analytics to choose a database from a named profile, or use init --no-login followed by doctor to reuse the default configuration directly. Profile login stores only the profile/config selector and connection details, keeping the token in .astrarc. The npm package name stays unchanged for compatibility.
Commands
As an MCP server
The 18 tools:
- Connect:
connection_status,list_databases - Explore:
database_overview,describe_collection,describe_table,list_vectorize_providers - Query:
find,vector_search,count,distinct_values - Build:
code_examples - Change:
insert,update,delete,create_collection,create_table,create_index,drop
Destructive operations need the user's confirmation, through elicitation or an explicit confirm argument after the user approves. Credentials come from the environment, then the project's .env, host settings, login --global, and the Astra CLI's ~/.astrarc.
The package bundles everything into one file with no runtime dependencies. Releases after 2.0.0 are published from GitHub Actions with npm provenance.
Documentation, the Claude Code and Codex plugin, and the hosted OAuth endpoint for ChatGPT and claude.ai: github.com/jestatsio/astra-db-plugin.
Apache-2.0 for this plugin's own code. Includes documentation-derived examples vendored from sl-at-ibm/astra-toolkit-skill; see the bundled NOTICE for attribution and upstream licensing status. Community project, not an official DataStax or IBM product.
Source: server/README.md at commit da4660c
Tools
0Version history
1- v2.0.0LatestSep 24, 2026
Score
85% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts