codebase-retrieval-context-engine 2.0.2__tar.gz → 2.0.3__tar.gz

{codebase_retrieval_context_engine-2.0.2 → codebase_retrieval_context_engine-2.0.3}/.env.example

@@ -1,104 +1,121 @@
-# ============================================================
-# Corbell — Environment Variables
-# Copy this file to .env and fill in the values you need.
-# ============================================================
-
-# ----------------------------------------------------------
-# LLM Provider API Keys (pick one or more)
-# ----------------------------------------------------------
-ANTHROPIC_API_KEY=
-OPENAI_API_KEY=
-GOOGLE_API_KEY=
-# Multiple Google keys for round-robin: GOOGLE_API_KEY=key1,key2,key3
-VOYAGE_API_KEY=
-# Multiple Voyage keys for round-robin: VOYAGE_API_KEY=pa-key1,pa-key2,pa-key3
-
-# AWS Bedrock
-BEDROCK_API_KEY=
-AWS_ACCESS_KEY_ID=
-AWS_SECRET_ACCESS_KEY=
-AWS_REGION=us-east-1
-
-# Azure OpenAI
-AZURE_OPENAI_API_KEY=
-AZURE_OPENAI_ENDPOINT=
-AZURE_OPENAI_DEPLOYMENT=
-AZURE_OPENAI_API_VERSION=2024-02-01
-
-# GCP Vertex AI
-GOOGLE_APPLICATION_CREDENTIALS=
-GCP_PROJECT=
-GOOGLE_CLOUD_PROJECT=
-GCP_REGION=us-central1
-
-# Generic fallback key (used when provider-specific key is not set)
-CORBELL_LLM_API_KEY=
-
-# ----------------------------------------------------------
-# Model Overrides
-# ----------------------------------------------------------
-
-# Embedding model — cloud only (default: voyage-code-3)
-# Supported: voyage-code-3, voyage-4-lite (VOYAGE_API_KEY), gemini-embedding-001 (GOOGLE_API_KEY)
-CORBELL_EMBEDDING_MODEL=
-
-# Embedding dimension override — overrides the model's default dimension.
-# Voyage default: 1024, Google (gemini-embedding-001) default: 768.
-CORBELL_EMBEDDING_DIM=
-
-# LLM model — generic (overrides llm.model default)
-CORBELL_LLM_MODEL=
-
-# LLM model — provider-specific (takes priority over CORBELL_LLM_MODEL)
-ANTHROPIC_MODEL=
-OPENAI_MODEL=
-GOOGLE_MODEL=
-OLLAMA_MODEL=
-AWS_MODEL=
-AZURE_MODEL=
-GCP_MODEL=
-
-# ----------------------------------------------------------
-# Workspace Path
-# ----------------------------------------------------------
-
-# Full path to the workspace (repository) root directory.
-# Used as fallback when --workspace-full-path is not passed to CLI.
-# Resolution order: --workspace-full-path flag → CORBELL_WORKSPACE → cwd
-CORBELL_WORKSPACE=
-
-# ----------------------------------------------------------
-# Query Configuration (all optional, sensible defaults shown)
-# ----------------------------------------------------------
-
-# Maximum number of code chunks returned per query (default: 50)
-CORBELL_TOP_K=
-
-# Enable LLM reranking of results (default: true)
-CORBELL_RERANK=
-
-# BFS depth for call-graph expansion (default: 2)
-CORBELL_EXPAND_CALL_DEPTH=
-
-# Maximum chunks added via graph expansion (default: 30)
-CORBELL_EXPAND_MAX_CHUNKS=
-
-# ----------------------------------------------------------
-# Indexing Configuration (all optional, sensible defaults shown)
-# ----------------------------------------------------------
-
-# Number of lines per embedding chunk (default: 50)
-CORBELL_CHUNK_SIZE=
-
-# Overlap between consecutive chunks in lines (default: 10)
-CORBELL_CHUNK_OVERLAP=
-
-# Maximum file size to index in bytes (default: 1048576 = 1 MB)
-CORBELL_MAX_FILE_BYTES=
-
-# Comma-separated list of directory names to skip during indexing
-# (in addition to built-in skip list: .git, node_modules, __pycache__, etc.)
-CORBELL_SKIP_DIRS=
-
-# Number of parallel workers for indexing (default: min(cpu_count, 8))
-# CORBELL_INDEX_WORKERS=4
+# ============================================================
+# Corbell — Environment Variables
+# Copy this file to .env and fill in the values you need.
+# ============================================================
+
+# ----------------------------------------------------------
+# LLM Provider API Keys (pick one or more)
+# ----------------------------------------------------------
+ANTHROPIC_API_KEY=
+OPENAI_API_KEY=
+GOOGLE_API_KEY=
+# Multiple Google keys for round-robin: GOOGLE_API_KEY=key1,key2,key3
+VOYAGE_API_KEY=
+# Multiple Voyage keys for round-robin: VOYAGE_API_KEY=pa-key1,pa-key2,pa-key3
+
+# AWS Bedrock
+BEDROCK_API_KEY=
+AWS_ACCESS_KEY_ID=
+AWS_SECRET_ACCESS_KEY=
+AWS_REGION=us-east-1
+
+# Azure OpenAI
+AZURE_OPENAI_API_KEY=
+AZURE_OPENAI_ENDPOINT=
+AZURE_OPENAI_DEPLOYMENT=
+AZURE_OPENAI_API_VERSION=2024-02-01
+
+# GCP Vertex AI
+GOOGLE_APPLICATION_CREDENTIALS=
+GCP_PROJECT=
+GOOGLE_CLOUD_PROJECT=
+GCP_REGION=us-central1
+
+# Generic fallback key (used when provider-specific key is not set)
+CORBELL_LLM_API_KEY=
+
+# ----------------------------------------------------------
+# Model Overrides
+# ----------------------------------------------------------
+
+# Embedding model — cloud only (default: voyage-code-3)
+# Supported: voyage-code-3, voyage-4-lite (VOYAGE_API_KEY), gemini-embedding-001 (GOOGLE_API_KEY)
+CORBELL_EMBEDDING_MODEL=
+
+# Embedding dimension override — overrides the model's default dimension.
+# Voyage default: 1024, Google (gemini-embedding-001) default: 768.
+CORBELL_EMBEDDING_DIM=
+
+# LLM model — generic (overrides llm.model default)
+CORBELL_LLM_MODEL=
+
+# LLM provider — which API to use for reranking
+# Options: anthropic, openai, google, ollama, aws, azure, gcp (default: anthropic)
+CORBELL_LLM_PROVIDER=
+
+# LLM model — provider-specific (takes priority over CORBELL_LLM_MODEL)
+ANTHROPIC_MODEL=
+OPENAI_MODEL=
+GOOGLE_MODEL=
+OLLAMA_MODEL=
+AWS_MODEL=
+AZURE_MODEL=
+GCP_MODEL=
+
+# ----------------------------------------------------------
+# Workspace Path
+# ----------------------------------------------------------
+
+# Full path to the workspace (repository) root directory.
+# Used as fallback when --workspace-full-path is not passed to CLI.
+# Resolution order: --workspace-full-path flag → CORBELL_WORKSPACE → cwd
+CORBELL_WORKSPACE=
+
+# ----------------------------------------------------------
+# Query Configuration (all optional, sensible defaults shown)
+# ----------------------------------------------------------
+
+# Maximum number of code chunks returned per query (default: 50)
+CORBELL_TOP_K=
+
+# Enable LLM reranking of results (default: true)
+CORBELL_RERANK=
+
+# BFS depth for call-graph expansion (default: 2)
+CORBELL_EXPAND_CALL_DEPTH=
+
+# Maximum chunks added via graph expansion (default: 30)
+CORBELL_EXPAND_MAX_CHUNKS=
+
+# ----------------------------------------------------------
+# Indexing Configuration (all optional, sensible defaults shown)
+# ----------------------------------------------------------
+
+# Number of lines per embedding chunk (default: 50)
+CORBELL_CHUNK_SIZE=
+
+# Overlap between consecutive chunks in lines (default: 10)
+CORBELL_CHUNK_OVERLAP=
+
+# Maximum file size to index in bytes (default: 1048576 = 1 MB)
+CORBELL_MAX_FILE_BYTES=
+
+# Comma-separated list of directory names to skip during indexing
+# (in addition to built-in skip list: .git, node_modules, __pycache__, etc.)
+CORBELL_SKIP_DIRS=
+
+# Number of parallel workers for indexing (default: min(cpu_count, 8))
+# CORBELL_INDEX_WORKERS=4
+
+# Number of concurrent embedding API threads per encode call (default: provider-aware)
+# VoyageEmbeddingModel default: 30, GoogleEmbeddingModel default: 8, other: 4
+CORBELL_EMBED_CONCURRENCY=
+
+# Super-batch size: number of chunks encoded and written per streaming slice (default: 6000)
+# Sized to saturate the concurrent thread pool. Lower values reduce peak memory
+# at the cost of throughput; higher values may exceed API rate limits.
+CORBELL_EMBED_BATCH=
+
+# Enable verbose performance logging during index builds (any non-empty value)
+# Also available as --verbose / -v flag on `corbell index build`
+CORBELL_VERBOSE=

codebase_retrieval_context_engine-2.0.3/PKG-INFO

@@ -0,0 +1,107 @@
+Metadata-Version: 2.4
+Name: codebase-retrieval-context-engine
+Version: 2.0.3
+Summary: Code retrieval engine — hybrid embedding + graph search for LLM context injection.
+Project-URL: Homepage, https://github.com/nullmastermind/local-context-engine
+Project-URL: Repository, https://github.com/nullmastermind/local-context-engine
+Project-URL: Issues, https://github.com/nullmastermind/local-context-engine/issues
+Author: nullmastermind
+License: Apache-2.0
+License-File: LICENSE
+Keywords: LLM,MCP,code-search,codebase-retrieval,context,embeddings,retrieval
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Software Development :: Libraries
+Requires-Python: >=3.11
+Requires-Dist: mcp>=1.1.2
+Requires-Dist: numpy>=2.0
+Requires-Dist: pathspec>=0.11
+Requires-Dist: pydantic>=2.0
+Requires-Dist: python-dotenv>=1.0
+Requires-Dist: rich>=13.0
+Requires-Dist: typer>=0.12
+Provides-Extra: anthropic
+Requires-Dist: anthropic>=0.25; extra == 'anthropic'
+Provides-Extra: aws
+Requires-Dist: boto3>=1.34; extra == 'aws'
+Provides-Extra: azure
+Requires-Dist: openai>=1.0; extra == 'azure'
+Provides-Extra: debug
+Requires-Dist: gradio>=4.0; extra == 'debug'
+Provides-Extra: dev
+Requires-Dist: httpx; extra == 'dev'
+Requires-Dist: mypy; extra == 'dev'
+Requires-Dist: pytest-asyncio; extra == 'dev'
+Requires-Dist: pytest-cov; extra == 'dev'
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: respx; extra == 'dev'
+Requires-Dist: ruff; extra == 'dev'
+Provides-Extra: gcp
+Requires-Dist: anthropic[vertex]>=0.25; extra == 'gcp'
+Requires-Dist: google-cloud-aiplatform>=1.38; extra == 'gcp'
+Provides-Extra: google
+Requires-Dist: google-genai>=2.7.0; extra == 'google'
+Provides-Extra: openai
+Requires-Dist: openai>=1.0; extra == 'openai'
+Provides-Extra: treesitter
+Requires-Dist: tree-sitter-c-sharp>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-go>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-java>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-javascript>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-php>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-python>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-ruby>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-rust>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter-typescript>=0.21; extra == 'treesitter'
+Requires-Dist: tree-sitter>=0.21; extra == 'treesitter'
+Provides-Extra: voyage
+Requires-Dist: voyageai>=0.3; extra == 'voyage'
+Description-Content-Type: text/markdown
+
+<div align="center">
+  <h1>codebase-retrieval-context-engine</h1>
+  <p><strong>Code retrieval engine for LLM context via MCP.</strong></p>
+  <p>
+    <a href="LICENSE"><img src="https://img.shields.io/badge/License-Apache_2.0-blue.svg" alt="License"/></a>
+  </p>
+</div>
+
+---
+
+## Add to Claude Code
+
+```bash
+claude mcp add codebase-retrieval -e CORBELL_LLM_PROVIDER=google -e GOOGLE_API_KEY=your-google-api-key -e GOOGLE_MODEL=gemini-3.1-flash-lite -e CORBELL_EMBEDDING_MODEL=voyage-4-lite -e VOYAGE_API_KEY=your-voyage-api-key -- uvx codebase-retrieval-context-engine
+```
+
+That's it. The AI agent passes workspace path and triggers index builds automatically.
+
+---
+
+## Build index manually (optional)
+
+```bash
+uvx codebase-retrieval-context-engine index build
+```
+
+Run from your project root. Env vars (`CORBELL_LLM_PROVIDER`, `GOOGLE_API_KEY`, etc.) must be set in your shell.
+
+---
+
+## Environment variables
+
+| Variable | Description |
+|---|---|
+| `CORBELL_LLM_PROVIDER` | LLM provider for reranking (`google`, `anthropic`, `openai`) |
+| `GOOGLE_API_KEY` | Google AI API key (supports multiple: `key1,key2,key3`) |
+| `GOOGLE_MODEL` | e.g. `gemini-3.1-flash-lite` |
+| `CORBELL_EMBEDDING_MODEL` | `voyage-4-lite`, `voyage-code-3`, or `gemini-embedding-001` |
+| `VOYAGE_API_KEY` | Voyage AI API key (supports multiple: `key1,key2,key3`). Add a card to billing to unlock rate limits. |
+
+---
+
+## License
+
+Apache 2.0

codebase_retrieval_context_engine-2.0.3/README.md

@@ -0,0 +1,45 @@
+<div align="center">
+  <h1>codebase-retrieval-context-engine</h1>
+  <p><strong>Code retrieval engine for LLM context via MCP.</strong></p>
+  <p>
+    <a href="LICENSE"><img src="https://img.shields.io/badge/License-Apache_2.0-blue.svg" alt="License"/></a>
+  </p>
+</div>
+
+---
+
+## Add to Claude Code
+
+```bash
+claude mcp add codebase-retrieval -e CORBELL_LLM_PROVIDER=google -e GOOGLE_API_KEY=your-google-api-key -e GOOGLE_MODEL=gemini-3.1-flash-lite -e CORBELL_EMBEDDING_MODEL=voyage-4-lite -e VOYAGE_API_KEY=your-voyage-api-key -- uvx codebase-retrieval-context-engine
+```
+
+That's it. The AI agent passes workspace path and triggers index builds automatically.
+
+---
+
+## Build index manually (optional)
+
+```bash
+uvx codebase-retrieval-context-engine index build
+```
+
+Run from your project root. Env vars (`CORBELL_LLM_PROVIDER`, `GOOGLE_API_KEY`, etc.) must be set in your shell.
+
+---
+
+## Environment variables
+
+| Variable | Description |
+|---|---|
+| `CORBELL_LLM_PROVIDER` | LLM provider for reranking (`google`, `anthropic`, `openai`) |
+| `GOOGLE_API_KEY` | Google AI API key (supports multiple: `key1,key2,key3`) |
+| `GOOGLE_MODEL` | e.g. `gemini-3.1-flash-lite` |
+| `CORBELL_EMBEDDING_MODEL` | `voyage-4-lite`, `voyage-code-3`, or `gemini-embedding-001` |
+| `VOYAGE_API_KEY` | Voyage AI API key (supports multiple: `key1,key2,key3`). Add a card to billing to unlock rate limits. |
+
+---
+
+## License
+
+Apache 2.0

{codebase_retrieval_context_engine-2.0.2 → codebase_retrieval_context_engine-2.0.3}/corbell/__init__.py

@@ -2,5 +2,5 @@
 Corbell — Code retrieval engine for LLM context injection.
 """
 
-__version__ = "2.0.2"
+__version__ = "2.0.3"
 __all__ = ["__version__"]