code-graph-rag 0.1.90__tar.gz → 0.2.10__tar.gz

{code_graph_rag-0.1.90/code_graph_rag.egg-info → code_graph_rag-0.2.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-graph-rag
-Version: 0.1.90
+Version: 0.2.10
 Summary: The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs
 License-Expression: MIT
 Keywords: rag,retrieval-augmented-generation,knowledge-graph,code-analysis,tree-sitter,mcp,mcp-server,llm,graph-database,semantic-search,codebase,memgraph,developer-tools,monorepo
@@ -19,10 +19,9 @@ Requires-Python: >=3.12
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: loguru>=0.7.3
-Requires-Dist: mcp<2,>=1.28.1
+Requires-Dist: mcp<2,>=1.30.0
 Requires-Dist: pydantic-ai-slim[anthropic,google,openai]>=2.0.0
 Requires-Dist: pydantic-settings>=2.12.0
-Requires-Dist: pymgclient>=1.6.0
 Requires-Dist: python-dotenv>=1.2.1
 Requires-Dist: tiktoken>=0.12.0
 Requires-Dist: tomli-w>=1.2.0
@@ -39,6 +38,8 @@ Requires-Dist: defusedxml>=0.7.1
 Requires-Dist: huggingface-hub[hf-xet]>=1.7.2
 Requires-Dist: pathspec>=1.0.4
 Requires-Dist: pygments>=2.20.0
+Requires-Dist: regex>=2024.11.6
+Requires-Dist: cgr-graph-rs<0.2,>=0.1.0; platform_python_implementation == "CPython" and ((sys_platform == "linux" and (platform_machine == "x86_64" or platform_machine == "aarch64")) or sys_platform == "darwin" or (sys_platform == "win32" and platform_machine == "AMD64"))
 Provides-Extra: test
 Requires-Dist: jedi>=0.19; extra == "test"
 Requires-Dist: pytest>=8.4.1; extra == "test"
@@ -50,8 +51,12 @@ Requires-Dist: libclang>=18.1.1; extra == "test"
 Requires-Dist: filelock>=3.12; extra == "test"
 Requires-Dist: ast-grep-py>=0.44.0; extra == "test"
 Requires-Dist: pyyaml>=6.0; extra == "test"
+Requires-Dist: pymgclient>=1.6.0; extra == "test"
+Requires-Dist: sqlite-vec>=0.1.9; extra == "test"
 Provides-Extra: cpp
 Requires-Dist: libclang>=18.1.1; extra == "cpp"
+Provides-Extra: memgraph
+Requires-Dist: pymgclient>=1.6.0; extra == "memgraph"
 Provides-Extra: neo4j
 Requires-Dist: neo4j>=5.28.1; extra == "neo4j"
 Provides-Extra: treesitter-full
@@ -77,6 +82,8 @@ Requires-Dist: torch>=2.6.0; extra == "semantic"
 Requires-Dist: transformers>=5.0.0; extra == "semantic"
 Provides-Extra: milvus
 Requires-Dist: pymilvus[milvus-lite]>=3.0.0; extra == "milvus"
+Provides-Extra: sqlite-vec
+Requires-Dist: sqlite-vec>=0.1.9; extra == "sqlite-vec"
 Provides-Extra: ast-grep
 Requires-Dist: ast-grep-py>=0.44.0; extra == "ast-grep"
 Requires-Dist: pyyaml>=6.0; extra == "ast-grep"
@@ -86,7 +93,7 @@ Dynamic: license-file
 
 # Code-Graph-RAG
 
-Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in Memgraph, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
+Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in an embedded graph database that needs no server, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
 
 ## What It Does
 
@@ -131,7 +138,9 @@ pip install 'code-graph-rag[semantic]'
 Qdrant is the default vector store for semantic search. To use Milvus Lite,
 install `code-graph-rag[semantic,milvus]`, then set
 `CGR_VECTOR_STORE_BACKEND=milvus` and `MILVUS_URI=./.milvus_code_embeddings.db`
-before indexing.
+before indexing. To keep the vectors in a SQLite file beside the embedded graph
+(`~/.cgr/vectors.sqlite`, no server), install `code-graph-rag[semantic,sqlite-vec]`
+and set `CGR_VECTOR_STORE_BACKEND=sqlite`.
 
 To compute embeddings on an OpenAI-compatible endpoint (OpenAI, Ollama, vLLM)
 instead of locally, set `CGR_EMBEDDING_PROVIDER=openai` with
@@ -166,14 +175,22 @@ for git installs and the other methods.
 ### Prerequisites
 
 - Python 3.12+
-- Docker (for Memgraph)
-- `cmake` (for building pymgclient)
 - `ripgrep` (`rg`) (for shell command text searching)
+- Only for the optional Memgraph backend: Docker, and `cmake` where no
+  `pymgclient` wheel fits your platform
+
+The knowledge graph is an embedded SQLite file (`~/.cgr/graph.sqlite`), so there
+is no database server to run. Memgraph and Neo4j remain available as opt-in
+backends; see [Graph Backends](https://docs.code-graph-rag.com/architecture/embedded-graph/).
 
 The wheel is pure Python (`py3-none-any`), so it installs on any platform with
-Python 3.12 or newer. Some distributions ship an older interpreter (Debian
-Bookworm, and so Raspberry Pi OS Bookworm, ships 3.11), which also explains the
-failed piwheels Bookworm build. On those systems, see the [installation
+Python 3.12 or newer. On Linux (x86_64, aarch64), macOS and Windows x64 it also
+pulls in `cgr-graph-rs`, a native extension that speeds up the embedded graph;
+elsewhere the same engine runs in pure Python.
+
+Some distributions ship an older interpreter (Debian Bookworm, and so Raspberry
+Pi OS Bookworm, ships 3.11), which also explains the failed piwheels Bookworm
+build. On those systems, see the [installation
 guide](https://github.com/vitali87/code-graph-rag/blob/main/docs/getting-started/installation.md#older-system-interpreters)
 for the available interpreter-pinning alternatives.
 
@@ -181,14 +198,17 @@ for the available interpreter-pinning alternatives.
 
 The package installs a `cgr` command.
 
-**Start Memgraph, parse a repo, and query it:**
+**Parse a repo, then query it:**
 
 ```bash
-cgr daemon up                              # start Memgraph + Qdrant
-cgr start --repo-path ./my-project \
-          --update-graph --clean           # parse & launch interactive chat
+cgr start --repo-path ./my-project --update-graph   # parse into the graph
+cgr start --repo-path ./my-project                  # launch the interactive chat
 ```
 
+The graph is written to `~/.cgr/graph.sqlite`. To use Memgraph instead, install
+`code-graph-rag[memgraph]`, set `GRAPH_BACKEND=memgraph`, and start the packaged
+Memgraph + Qdrant stack with `cgr daemon up`.
+
 **Index to protobuf for offline use:**
 
 ```bash
@@ -295,7 +315,11 @@ for fn in functions[:5]:
     print(f"{fn.properties['name']}: {len(rels)} relationships")
 ```
 
-### Query Memgraph with Cypher
+### Query the graph with Cypher
+
+`MemgraphIngestor` keeps its original name but works with whichever backend
+`GRAPH_BACKEND` selects. On the default embedded graph, `host` and `port` are
+not used.
 
 ```python
 from cgr import MemgraphIngestor
@@ -348,8 +372,10 @@ Configure via `.env` or environment variables:
 
 | Variable | Default | Description |
 |----------|---------|-------------|
-| `MEMGRAPH_HOST` | `localhost` | Memgraph hostname |
-| `MEMGRAPH_PORT` | `7687` | Memgraph port |
+| `GRAPH_BACKEND` | `sqlite` | Graph engine: `sqlite` (embedded), `memgraph`, or `neo4j` |
+| `SQLITE_GRAPH_PATH` | `~/.cgr/graph.sqlite` | Embedded graph file |
+| `MEMGRAPH_HOST` | `localhost` | Memgraph hostname (with `GRAPH_BACKEND=memgraph`) |
+| `MEMGRAPH_PORT` | `7687` | Memgraph port (with `GRAPH_BACKEND=memgraph`) |
 | `ORCHESTRATOR_PROVIDER` | | Provider: `google`, `openai`, `anthropic`, `azure`, `ollama`, `minimax`, `litellm_proxy` |
 | `ORCHESTRATOR_MODEL` | | Model ID (e.g. `gpt-5.6-terra`, `gemini-3.6-flash`, `claude-sonnet-5`, `qwen2.5-coder`) |
 | `ORCHESTRATOR_API_KEY` | | API key for the provider (not needed for `ollama`) |

{code_graph_rag-0.1.90 → code_graph_rag-0.2.10}/PYPI_README.md

@@ -1,6 +1,6 @@
 # Code-Graph-RAG
 
-Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in Memgraph, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
+Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in an embedded graph database that needs no server, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
 
 ## What It Does
 
@@ -45,7 +45,9 @@ pip install 'code-graph-rag[semantic]'
 Qdrant is the default vector store for semantic search. To use Milvus Lite,
 install `code-graph-rag[semantic,milvus]`, then set
 `CGR_VECTOR_STORE_BACKEND=milvus` and `MILVUS_URI=./.milvus_code_embeddings.db`
-before indexing.
+before indexing. To keep the vectors in a SQLite file beside the embedded graph
+(`~/.cgr/vectors.sqlite`, no server), install `code-graph-rag[semantic,sqlite-vec]`
+and set `CGR_VECTOR_STORE_BACKEND=sqlite`.
 
 To compute embeddings on an OpenAI-compatible endpoint (OpenAI, Ollama, vLLM)
 instead of locally, set `CGR_EMBEDDING_PROVIDER=openai` with
@@ -80,14 +82,22 @@ for git installs and the other methods.
 ### Prerequisites
 
 - Python 3.12+
-- Docker (for Memgraph)
-- `cmake` (for building pymgclient)
 - `ripgrep` (`rg`) (for shell command text searching)
+- Only for the optional Memgraph backend: Docker, and `cmake` where no
+  `pymgclient` wheel fits your platform
+
+The knowledge graph is an embedded SQLite file (`~/.cgr/graph.sqlite`), so there
+is no database server to run. Memgraph and Neo4j remain available as opt-in
+backends; see [Graph Backends](https://docs.code-graph-rag.com/architecture/embedded-graph/).
 
 The wheel is pure Python (`py3-none-any`), so it installs on any platform with
-Python 3.12 or newer. Some distributions ship an older interpreter (Debian
-Bookworm, and so Raspberry Pi OS Bookworm, ships 3.11), which also explains the
-failed piwheels Bookworm build. On those systems, see the [installation
+Python 3.12 or newer. On Linux (x86_64, aarch64), macOS and Windows x64 it also
+pulls in `cgr-graph-rs`, a native extension that speeds up the embedded graph;
+elsewhere the same engine runs in pure Python.
+
+Some distributions ship an older interpreter (Debian Bookworm, and so Raspberry
+Pi OS Bookworm, ships 3.11), which also explains the failed piwheels Bookworm
+build. On those systems, see the [installation
 guide](https://github.com/vitali87/code-graph-rag/blob/main/docs/getting-started/installation.md#older-system-interpreters)
 for the available interpreter-pinning alternatives.
 
@@ -95,14 +105,17 @@ for the available interpreter-pinning alternatives.
 
 The package installs a `cgr` command.
 
-**Start Memgraph, parse a repo, and query it:**
+**Parse a repo, then query it:**
 
 ```bash
-cgr daemon up                              # start Memgraph + Qdrant
-cgr start --repo-path ./my-project \
-          --update-graph --clean           # parse & launch interactive chat
+cgr start --repo-path ./my-project --update-graph   # parse into the graph
+cgr start --repo-path ./my-project                  # launch the interactive chat
 ```
 
+The graph is written to `~/.cgr/graph.sqlite`. To use Memgraph instead, install
+`code-graph-rag[memgraph]`, set `GRAPH_BACKEND=memgraph`, and start the packaged
+Memgraph + Qdrant stack with `cgr daemon up`.
+
 **Index to protobuf for offline use:**
 
 ```bash
@@ -209,7 +222,11 @@ for fn in functions[:5]:
     print(f"{fn.properties['name']}: {len(rels)} relationships")
 ```
 
-### Query Memgraph with Cypher
+### Query the graph with Cypher
+
+`MemgraphIngestor` keeps its original name but works with whichever backend
+`GRAPH_BACKEND` selects. On the default embedded graph, `host` and `port` are
+not used.
 
 ```python
 from cgr import MemgraphIngestor
@@ -262,8 +279,10 @@ Configure via `.env` or environment variables:
 
 | Variable | Default | Description |
 |----------|---------|-------------|
-| `MEMGRAPH_HOST` | `localhost` | Memgraph hostname |
-| `MEMGRAPH_PORT` | `7687` | Memgraph port |
+| `GRAPH_BACKEND` | `sqlite` | Graph engine: `sqlite` (embedded), `memgraph`, or `neo4j` |
+| `SQLITE_GRAPH_PATH` | `~/.cgr/graph.sqlite` | Embedded graph file |
+| `MEMGRAPH_HOST` | `localhost` | Memgraph hostname (with `GRAPH_BACKEND=memgraph`) |
+| `MEMGRAPH_PORT` | `7687` | Memgraph port (with `GRAPH_BACKEND=memgraph`) |
 | `ORCHESTRATOR_PROVIDER` | | Provider: `google`, `openai`, `anthropic`, `azure`, `ollama`, `minimax`, `litellm_proxy` |
 | `ORCHESTRATOR_MODEL` | | Model ID (e.g. `gpt-5.6-terra`, `gemini-3.6-flash`, `claude-sonnet-5`, `qwen2.5-coder`) |
 | `ORCHESTRATOR_API_KEY` | | API key for the provider (not needed for `ollama`) |

{code_graph_rag-0.1.90 → code_graph_rag-0.2.10}/README.md

@@ -42,13 +42,13 @@
 
 # Code-Graph-RAG
 
-Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in Memgraph, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
+Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in an embedded graph database that needs no server, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
 
 <p align="center">
   <b>1. Index:</b> <code>cgr start --update-graph</code> parses a repository into a knowledge graph (sped up)
 </p>
 <p align="center">
-  <img src="./assets/demo-indexing.gif" alt="cgr parsing the code-graph-rag repository into a Memgraph knowledge graph, then printing node and relationship counts">
+  <img src="./assets/demo-indexing.gif" alt="cgr parsing the code-graph-rag repository into a knowledge graph, then printing node and relationship counts">
 </p>
 
 <p align="center">
@@ -61,8 +61,7 @@ Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by c
 ## Latest News 🔥
 
 <!-- SECTION:latest_news -->
-- **Contextual Queries**: Support for more detailed context queries has been added, including reasons and evaluation ingestor reads, enhancing the understanding of code interactions.
-- **Graph Improvements**: The graph analysis capabilities have been expanded to follow cross-service links and implementors/overrides, providing a more comprehensive view of code dependencies.
+- **GitHub Sponsors**: Added support for GitHub Sponsors via the command line.
 <!-- /SECTION:latest_news -->
 
 See [NEWS.md](NEWS.md) for the full history.
@@ -83,15 +82,17 @@ Point Code-Graph-RAG at a repository and it reads every source file, extracts fu
 
 The system has two components:
 
-1. **Multi-language parser.** A Tree-sitter based parser reads the codebase and ingests functions, classes, methods, modules, and their relationships into Memgraph under a single language-agnostic schema. Where a toolchain is available, compiler-grade frontends layer exact facts on top (libclang for C/C++, `go/types` for Go, and opt-in Roslyn, `javac` and Jedi for C#, Java and Python), and [dynamic tracing](docs/guide/dynamic-tracing.md) merges calls observed at runtime. Tree-sitter stays the backbone: a trace only sees code that ran, and a compiler frontend only covers what its toolchain can build.
+1. **Multi-language parser.** A Tree-sitter based parser reads the codebase and ingests functions, classes, methods, modules, and their relationships into the graph under a single language-agnostic schema. Where a toolchain is available, compiler-grade frontends layer exact facts on top (libclang for C/C++, `go/types` for Go, and opt-in Roslyn, `javac` and Jedi for C#, Java and Python), and [dynamic tracing](docs/guide/dynamic-tracing.md) merges calls observed at runtime. Tree-sitter stays the backbone: a trace only sees code that ran, and a compiler frontend only covers what its toolchain can build.
 2. **RAG system** (`codebase_rag/`). An interactive CLI that turns natural language into Cypher queries, retrieves matching code, and drives AI-powered editing and optimisation.
 
 ```
-Source Code -> Tree-sitter Parser -> AST Analysis -> Memgraph Knowledge Graph
+Source Code -> Tree-sitter Parser -> AST Analysis -> Knowledge Graph
                                                              |
 User Query -> AI Model (Cypher Gen) -> Cypher Query -> Graph Results -> Response
 ```
 
+The graph is stored in an embedded SQLite file by default. Memgraph and Neo4j are opt-in backends that run the same Cypher; [Graph Backends](docs/architecture/embedded-graph.md) compares them.
+
 See the [Architecture Overview](docs/architecture/overview.md) and [Graph Schema](docs/architecture/graph-schema.md) for the full picture.
 
 ## Supported Languages
@@ -137,7 +138,7 @@ uv tool install "code-graph-rag[treesitter-full,ast-grep,semantic] @ git+https:/
 
 To upgrade an existing install, run `uv tool upgrade code-graph-rag` (with pipx, `pipx upgrade code-graph-rag`, or `pipx reinstall code-graph-rag` for a git install). [Upgrade](docs/getting-started/installation.md#upgrade) covers the other install methods.
 
-You also need Python 3.12+, Docker (for Memgraph), `cmake`, and `ripgrep`. Full prerequisites, source installs, and environment setup are in the [Installation](docs/getting-started/installation.md) guide.
+You also need Python 3.12+ and `ripgrep`. The graph lives in an embedded SQLite file by default, so there is no database server to run: no Docker and no `cmake`. Memgraph and Neo4j remain available as opt-in backends ([Graph Backends](docs/architecture/embedded-graph.md)). Full prerequisites, source installs, and environment setup are in the [Installation](docs/getting-started/installation.md) guide.
 
 > [!NOTE]
 > The wheel is pure Python (`py3-none-any`), so it installs on any platform with Python 3.12 or newer. Older system interpreters (Debian Bookworm ships 3.11) need the interpreter pinned explicitly; see [Installation](docs/getting-started/installation.md#older-system-interpreters) for the commands.
@@ -145,14 +146,16 @@ You also need Python 3.12+, Docker (for Memgraph), `cmake`, and `ripgrep`. Full
 ## Quick Start
 
 ```bash
-# Start the packaged Memgraph + Qdrant stack (no compose file needed)
-cgr daemon up
-
 # Parse a repository into the graph, then query it
 cgr start --repo-path /path/to/repo --update-graph
 cgr start --repo-path /path/to/repo
 ```
 
+The graph is written to `~/.cgr/graph.sqlite` (set `SQLITE_GRAPH_PATH` to move it).
+To use Memgraph instead, install the driver with the `memgraph` extra, set
+`GRAPH_BACKEND=memgraph`, and start the packaged Memgraph + Qdrant stack with
+`cgr daemon up`.
+
 Repeat the first command for each repository you want indexed; the graph is
 shared, and syncing one project leaves the others alone. To start over from an
 empty graph, add `--clean` — it deletes **every** project in the shared graph,
@@ -188,6 +191,7 @@ Code-Graph-RAG runs as an [MCP](https://modelcontextprotocol.io) server so Claud
 **Architecture**
 - [Overview](docs/architecture/overview.md)
 - [Graph Schema](docs/architecture/graph-schema.md)
+- [Graph Backends](docs/architecture/embedded-graph.md)
 - [Language Support](docs/architecture/language-support.md)
 - [Data-Flow Edges](docs/architecture/data-flow-edges.md)
 
@@ -222,6 +226,8 @@ Please see [CONTRIBUTING.md](CONTRIBUTING.md) for contribution guidelines. Good
 
 For issues or questions, check the [Troubleshooting](docs/advanced/troubleshooting.md) guide first, then open an issue.
 
+If cgr is useful to you, please consider pledging support on [GitHub Sponsors](https://github.com/sponsors/vitali87). cgr links the page under `cgr --version` and, in a terminal, after most commands finish; set `CGR_SPONSOR_PLEDGE=false` to hide that line.
+
 ## License
 
 MIT. See [LICENSE](LICENSE).

{code_graph_rag-0.1.90 → code_graph_rag-0.2.10/code_graph_rag.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-graph-rag
-Version: 0.1.90
+Version: 0.2.10
 Summary: The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs
 License-Expression: MIT
 Keywords: rag,retrieval-augmented-generation,knowledge-graph,code-analysis,tree-sitter,mcp,mcp-server,llm,graph-database,semantic-search,codebase,memgraph,developer-tools,monorepo
@@ -19,10 +19,9 @@ Requires-Python: >=3.12
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: loguru>=0.7.3
-Requires-Dist: mcp<2,>=1.28.1
+Requires-Dist: mcp<2,>=1.30.0
 Requires-Dist: pydantic-ai-slim[anthropic,google,openai]>=2.0.0
 Requires-Dist: pydantic-settings>=2.12.0
-Requires-Dist: pymgclient>=1.6.0
 Requires-Dist: python-dotenv>=1.2.1
 Requires-Dist: tiktoken>=0.12.0
 Requires-Dist: tomli-w>=1.2.0
@@ -39,6 +38,8 @@ Requires-Dist: defusedxml>=0.7.1
 Requires-Dist: huggingface-hub[hf-xet]>=1.7.2
 Requires-Dist: pathspec>=1.0.4
 Requires-Dist: pygments>=2.20.0
+Requires-Dist: regex>=2024.11.6
+Requires-Dist: cgr-graph-rs<0.2,>=0.1.0; platform_python_implementation == "CPython" and ((sys_platform == "linux" and (platform_machine == "x86_64" or platform_machine == "aarch64")) or sys_platform == "darwin" or (sys_platform == "win32" and platform_machine == "AMD64"))
 Provides-Extra: test
 Requires-Dist: jedi>=0.19; extra == "test"
 Requires-Dist: pytest>=8.4.1; extra == "test"
@@ -50,8 +51,12 @@ Requires-Dist: libclang>=18.1.1; extra == "test"
 Requires-Dist: filelock>=3.12; extra == "test"
 Requires-Dist: ast-grep-py>=0.44.0; extra == "test"
 Requires-Dist: pyyaml>=6.0; extra == "test"
+Requires-Dist: pymgclient>=1.6.0; extra == "test"
+Requires-Dist: sqlite-vec>=0.1.9; extra == "test"
 Provides-Extra: cpp
 Requires-Dist: libclang>=18.1.1; extra == "cpp"
+Provides-Extra: memgraph
+Requires-Dist: pymgclient>=1.6.0; extra == "memgraph"
 Provides-Extra: neo4j
 Requires-Dist: neo4j>=5.28.1; extra == "neo4j"
 Provides-Extra: treesitter-full
@@ -77,6 +82,8 @@ Requires-Dist: torch>=2.6.0; extra == "semantic"
 Requires-Dist: transformers>=5.0.0; extra == "semantic"
 Provides-Extra: milvus
 Requires-Dist: pymilvus[milvus-lite]>=3.0.0; extra == "milvus"
+Provides-Extra: sqlite-vec
+Requires-Dist: sqlite-vec>=0.1.9; extra == "sqlite-vec"
 Provides-Extra: ast-grep
 Requires-Dist: ast-grep-py>=0.44.0; extra == "ast-grep"
 Requires-Dist: pyyaml>=6.0; extra == "ast-grep"
@@ -86,7 +93,7 @@ Dynamic: license-file
 
 # Code-Graph-RAG
 
-Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in Memgraph, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
+Code-Graph-RAG parses a multi-language codebase with Tree-sitter, sharpened by compiler-grade frontends and runtime traces where available, builds a knowledge graph of its structure in an embedded graph database that needs no server, and lets you query, edit, and optimise that code in plain English. It works across a monorepo of mixed languages under one unified graph schema.
 
 ## What It Does
 
@@ -131,7 +138,9 @@ pip install 'code-graph-rag[semantic]'
 Qdrant is the default vector store for semantic search. To use Milvus Lite,
 install `code-graph-rag[semantic,milvus]`, then set
 `CGR_VECTOR_STORE_BACKEND=milvus` and `MILVUS_URI=./.milvus_code_embeddings.db`
-before indexing.
+before indexing. To keep the vectors in a SQLite file beside the embedded graph
+(`~/.cgr/vectors.sqlite`, no server), install `code-graph-rag[semantic,sqlite-vec]`
+and set `CGR_VECTOR_STORE_BACKEND=sqlite`.
 
 To compute embeddings on an OpenAI-compatible endpoint (OpenAI, Ollama, vLLM)
 instead of locally, set `CGR_EMBEDDING_PROVIDER=openai` with
@@ -166,14 +175,22 @@ for git installs and the other methods.
 ### Prerequisites
 
 - Python 3.12+
-- Docker (for Memgraph)
-- `cmake` (for building pymgclient)
 - `ripgrep` (`rg`) (for shell command text searching)
+- Only for the optional Memgraph backend: Docker, and `cmake` where no
+  `pymgclient` wheel fits your platform
+
+The knowledge graph is an embedded SQLite file (`~/.cgr/graph.sqlite`), so there
+is no database server to run. Memgraph and Neo4j remain available as opt-in
+backends; see [Graph Backends](https://docs.code-graph-rag.com/architecture/embedded-graph/).
 
 The wheel is pure Python (`py3-none-any`), so it installs on any platform with
-Python 3.12 or newer. Some distributions ship an older interpreter (Debian
-Bookworm, and so Raspberry Pi OS Bookworm, ships 3.11), which also explains the
-failed piwheels Bookworm build. On those systems, see the [installation
+Python 3.12 or newer. On Linux (x86_64, aarch64), macOS and Windows x64 it also
+pulls in `cgr-graph-rs`, a native extension that speeds up the embedded graph;
+elsewhere the same engine runs in pure Python.
+
+Some distributions ship an older interpreter (Debian Bookworm, and so Raspberry
+Pi OS Bookworm, ships 3.11), which also explains the failed piwheels Bookworm
+build. On those systems, see the [installation
 guide](https://github.com/vitali87/code-graph-rag/blob/main/docs/getting-started/installation.md#older-system-interpreters)
 for the available interpreter-pinning alternatives.
 
@@ -181,14 +198,17 @@ for the available interpreter-pinning alternatives.
 
 The package installs a `cgr` command.
 
-**Start Memgraph, parse a repo, and query it:**
+**Parse a repo, then query it:**
 
 ```bash
-cgr daemon up                              # start Memgraph + Qdrant
-cgr start --repo-path ./my-project \
-          --update-graph --clean           # parse & launch interactive chat
+cgr start --repo-path ./my-project --update-graph   # parse into the graph
+cgr start --repo-path ./my-project                  # launch the interactive chat
 ```
 
+The graph is written to `~/.cgr/graph.sqlite`. To use Memgraph instead, install
+`code-graph-rag[memgraph]`, set `GRAPH_BACKEND=memgraph`, and start the packaged
+Memgraph + Qdrant stack with `cgr daemon up`.
+
 **Index to protobuf for offline use:**
 
 ```bash
@@ -295,7 +315,11 @@ for fn in functions[:5]:
     print(f"{fn.properties['name']}: {len(rels)} relationships")
 ```
 
-### Query Memgraph with Cypher
+### Query the graph with Cypher
+
+`MemgraphIngestor` keeps its original name but works with whichever backend
+`GRAPH_BACKEND` selects. On the default embedded graph, `host` and `port` are
+not used.
 
 ```python
 from cgr import MemgraphIngestor
@@ -348,8 +372,10 @@ Configure via `.env` or environment variables:
 
 | Variable | Default | Description |
 |----------|---------|-------------|
-| `MEMGRAPH_HOST` | `localhost` | Memgraph hostname |
-| `MEMGRAPH_PORT` | `7687` | Memgraph port |
+| `GRAPH_BACKEND` | `sqlite` | Graph engine: `sqlite` (embedded), `memgraph`, or `neo4j` |
+| `SQLITE_GRAPH_PATH` | `~/.cgr/graph.sqlite` | Embedded graph file |
+| `MEMGRAPH_HOST` | `localhost` | Memgraph hostname (with `GRAPH_BACKEND=memgraph`) |
+| `MEMGRAPH_PORT` | `7687` | Memgraph port (with `GRAPH_BACKEND=memgraph`) |
 | `ORCHESTRATOR_PROVIDER` | | Provider: `google`, `openai`, `anthropic`, `azure`, `ollama`, `minimax`, `litellm_proxy` |
 | `ORCHESTRATOR_MODEL` | | Model ID (e.g. `gpt-5.6-terra`, `gemini-3.6-flash`, `claude-sonnet-5`, `qwen2.5-coder`) |
 | `ORCHESTRATOR_API_KEY` | | API key for the provider (not needed for `ollama`) |

{code_graph_rag-0.1.90 → code_graph_rag-0.2.10}/code_graph_rag.egg-info/SOURCES.txt

@@ -46,6 +46,7 @@ codebase_rag/interface_skip.py
 codebase_rag/language_spec.py
 codebase_rag/logs.py
 codebase_rag/main.py
+codebase_rag/mgclient_compat.py
 codebase_rag/mgclient_probe.py
 codebase_rag/models.py
 codebase_rag/parser_fingerprint.py
@@ -58,6 +59,7 @@ codebase_rag/schema_builder.py
 codebase_rag/schema_parse.py
 codebase_rag/schemas.py
 codebase_rag/sql_names.py
+codebase_rag/sqlite_vectors.py
 codebase_rag/structural_check.py
 codebase_rag/structural_delta.py
 codebase_rag/sync_guard.py
@@ -137,6 +139,7 @@ codebase_rag/constants/core.py
 codebase_rag/constants/deadcode_roots.py
 codebase_rag/constants/deps.py
 codebase_rag/constants/duplicates.py
+codebase_rag/constants/embedded_graph.py
 codebase_rag/constants/fqn_specs.py
 codebase_rag/constants/graph.py
 codebase_rag/constants/health.py
@@ -147,6 +150,7 @@ codebase_rag/constants/mcp.py
 codebase_rag/constants/notebook.py
 codebase_rag/constants/providers.py
 codebase_rag/constants/security.py
+codebase_rag/constants/sqlite_vectors.py
 codebase_rag/constants/stdlib_types.py
 codebase_rag/constants/structural.py
 codebase_rag/constants/trace.py
@@ -158,10 +162,27 @@ codebase_rag/editing/imports.py
 codebase_rag/editing/occurrences.py
 codebase_rag/editing/patcher.py
 codebase_rag/editing/rename.py
+codebase_rag/editing/signature_bind.py
 codebase_rag/editing/signature_contract.py
 codebase_rag/editing/signature_spec.py
 codebase_rag/editing/sites.py
 codebase_rag/editing/transaction.py
+codebase_rag/embedded_graph/__init__.py
+codebase_rag/embedded_graph/compiler.py
+codebase_rag/embedded_graph/connection.py
+codebase_rag/embedded_graph/cypher_ast.py
+codebase_rag/embedded_graph/executor.py
+codebase_rag/embedded_graph/expressions.py
+codebase_rag/embedded_graph/extension.py
+codebase_rag/embedded_graph/functions.py
+codebase_rag/embedded_graph/lexer.py
+codebase_rag/embedded_graph/limits.py
+codebase_rag/embedded_graph/native.py
+codebase_rag/embedded_graph/parser.py
+codebase_rag/embedded_graph/procedures.py
+codebase_rag/embedded_graph/sql.py
+codebase_rag/embedded_graph/store.py
+codebase_rag/embedded_graph/types.py
 codebase_rag/mcp/__init__.py
 codebase_rag/mcp/client.py
 codebase_rag/mcp/server.py

{code_graph_rag-0.1.90 → code_graph_rag-0.2.10}/code_graph_rag.egg-info/requires.txt

@@ -1,8 +1,7 @@
 loguru>=0.7.3
-mcp<2,>=1.28.1
+mcp<2,>=1.30.0
 pydantic-ai-slim[anthropic,google,openai]>=2.0.0
 pydantic-settings>=2.12.0
-pymgclient>=1.6.0
 python-dotenv>=1.2.1
 tiktoken>=0.12.0
 tomli-w>=1.2.0
@@ -19,6 +18,10 @@ defusedxml>=0.7.1
 huggingface-hub[hf-xet]>=1.7.2
 pathspec>=1.0.4
 pygments>=2.20.0
+regex>=2024.11.6
+
+[:platform_python_implementation == "CPython" and ((sys_platform == "linux" and (platform_machine == "x86_64" or platform_machine == "aarch64")) or sys_platform == "darwin" or (sys_platform == "win32" and platform_machine == "AMD64"))]
+cgr-graph-rs<0.2,>=0.1.0
 
 [ast-grep]
 ast-grep-py>=0.44.0
@@ -27,6 +30,9 @@ pyyaml>=6.0
 [cpp]
 libclang>=18.1.1
 
+[memgraph]
+pymgclient>=1.6.0
+
 [milvus]
 pymilvus[milvus-lite]>=3.0.0
 
@@ -41,6 +47,9 @@ qdrant-client>=1.9.0
 torch>=2.6.0
 transformers>=5.0.0
 
+[sqlite-vec]
+sqlite-vec>=0.1.9
+
 [test]
 jedi>=0.19
 pytest>=8.4.1
@@ -52,6 +61,8 @@ libclang>=18.1.1
 filelock>=3.12
 ast-grep-py>=0.44.0
 pyyaml>=6.0
+pymgclient>=1.6.0
+sqlite-vec>=0.1.9
 
 [treesitter-full]
 tree-sitter-python>=0.23.6