Skip to main content

dbx-tools-graphiti

Native launcher for Graphiti with a local Neo4j backend. It runs both services directly as host processes; it does not use Docker, Podman, or another container runtime.

Install from PyPI:

uv add dbx-tools-graphiti

Or install the current main branch:

uv add "dbx-tools-graphiti @ git+https://github.com/reggie-db/dbx-tools.git@main#subdirectory=packages/py/graphiti"

Key features

  • launches upstream Graphiti's HTTP MCP server at http://127.0.0.1:8000/mcp/;
  • runs Neo4j Community 5.26 as a native background process;
  • provisions Java 21 and uv through mise use -g only when absent;
  • pins Graphiti and Neo4j versions for repeatable local environments;
  • caches downloads, Python dependencies, Neo4j data, credentials, and logs;
  • supports foreground or background operation without vendoring Graphiti code.

Quick start

mise must already be installed. The launcher handles Java and uv itself.

export OPENAI_API_KEY=...
uv run dbx-graphiti start

The first run downloads about 120 MB of Neo4j plus the pinned Graphiti release, creates Graphiti's uv environment, generates a local Neo4j password, starts Neo4j, and then runs Graphiti in the foreground. Later runs reuse all of it.

For background operation:

uv run dbx-graphiti up
uv run dbx-graphiti status
uv run dbx-graphiti down

Commands

Command Behavior
setup Provision tools and populate the local cache without starting services.
start Start Neo4j, then run Graphiti in the foreground. This is the default.
up Start both services in the background.
down Stop the managed Graphiti and Neo4j processes.
status Print process state and the MCP URL as JSON.
env Print resolved Neo4j connection settings as JSON. Treat its output as secret.

Arguments after start or up are forwarded to upstream Graphiti. For example:

uv run dbx-graphiti start -- --port 9000 --group-id my-agent

Provisioning and caching

The package deliberately keeps orchestration separate from Graphiti itself:

  1. It checks mise where java@21 and mise where uv@0.11.
  2. A missing tool is installed globally with mise use -g --yes.
  3. Neo4j Community 5.26.12 is downloaded from dist.neo4j.org and unpacked.
  4. Graphiti v0.29.3 is downloaded from its GitHub release tag.
  5. uv sync --project <checkout>/mcp_server creates the upstream environment.
  6. A generated Neo4j password is stored with mode 0600 and supplied to Graphiti through its documented environment variables.

The cache root is:

  • macOS: ~/Library/Application Support/dbx-tools/graphiti
  • Linux: ${XDG_DATA_HOME:-~/.local/share}/dbx-tools/graphiti
  • Windows: %LOCALAPPDATA%/dbx-tools/graphiti

Set DBX_GRAPHITI_HOME to override it. Removing the directory clears the download cache and permanently removes the local graph data.

Configuration

The packaged default fixes the database provider to Neo4j and otherwise follows upstream Graphiti environment names. Common settings are:

Variable Default
OPENAI_API_KEY Required
MODEL_NAME gpt-4.1-mini
EMBEDDER_MODEL text-embedding-3-small
GRAPHITI_GROUP_ID main
GRAPHITI_HOST 127.0.0.1
GRAPHITI_PORT 8000
NEO4J_URI bolt://127.0.0.1:7687
NEO4J_DATABASE neo4j

Explicit NEO4J_* values override generated defaults, which lets the Graphiti process use an existing Neo4j server. The launcher still manages its local Neo4j process; use upstream Graphiti directly if lifecycle ownership belongs to an external database administrator.

Graphiti owns MCP tools, graph behavior, LLM calls, embeddings, and migrations. This package owns only repeatable installation, configuration, and process lifecycle. See the upstream MCP server documentation for its complete API and provider configuration.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dbx_tools_graphiti-0.6.111.tar.gz (6.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

dbx_tools_graphiti-0.6.111-py3-none-any.whl (9.0 kB view details)

Uploaded Python 3

File details

Details for the file dbx_tools_graphiti-0.6.111.tar.gz.

File metadata

  • Download URL: dbx_tools_graphiti-0.6.111.tar.gz
  • Upload date:
  • Size: 6.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for dbx_tools_graphiti-0.6.111.tar.gz
Algorithm Hash digest
SHA256 93f5519d36bca6a91b16f550f757fdd4abe1feb04af2d8a94e75b3641df2e529
MD5 f3f94f54358b8441912e24af057dcbe7
BLAKE2b-256 3325564e643ba4fcc06f9b5a8a166fc627ad3caa8bc500485212aa5cf213b697

See more details on using hashes here.

Provenance

The following attestation bundles were made for dbx_tools_graphiti-0.6.111.tar.gz:

Publisher: python-release.yml on reggie-db/dbx-tools

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file dbx_tools_graphiti-0.6.111-py3-none-any.whl.

File metadata

File hashes

Hashes for dbx_tools_graphiti-0.6.111-py3-none-any.whl
Algorithm Hash digest
SHA256 550e6bcbd9feb913726fd8a451666ab0619399af118baf4abe6df5a0e1896b4e
MD5 3a96a6fd194c1f1f9ab26b1eb3c65e18
BLAKE2b-256 1037b51638edec8beb28dc074b33167fc9d3287b4aa650ec0fede8a99daf4598

See more details on using hashes here.

Provenance

The following attestation bundles were made for dbx_tools_graphiti-0.6.111-py3-none-any.whl:

Publisher: python-release.yml on reggie-db/dbx-tools

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page