chore: removed unused config values

.env.example

@@ -8,15 +8,6 @@ POSTGRES_TABLE_NAME="documents"
 
 # General Configuration
 PORT="3001"
-SIMILARITY_MEASURE="cosine"
-
-# LLM Provider Configuration
-DEFAULT_CHAT_PROVIDER="gemini"
-DEFAULT_CHAT_MODEL="Gemini Flash 2.5"
-DEFAULT_FAST_CHAT_PROVIDER="gemini"
-DEFAULT_FAST_CHAT_MODEL="Gemini Flash 2.5"
-DEFAULT_EMBEDDING_PROVIDER="openai"
-DEFAULT_EMBEDDING_MODEL="Text embedding 3 large"
 
 # LLM Provider API Keys
 OPENAI_API_KEY=""
@@ -26,10 +17,6 @@ DEEPSEEK_API_KEY=""
 GROQ_API_KEY=""
 XAI_API_KEY=""
 
-# Version Configuration
-STARKNET_FOUNDRY_VERSION="0.47.0"
-SCARB_VERSION="2.11.4"
-
 # LangSmith Configuration (Optional)
 LANGSMITH_TRACING="false"
 LANGSMITH_ENDPOINT="https://api.smith.langchain.com"

.github/workflows/generate-embeddings.yml

@@ -64,24 +64,12 @@ jobs:
 
           # General Configuration
           PORT="3001"
-          SIMILARITY_MEASURE="cosine"
-
-          # Provider Configuration
-          DEFAULT_CHAT_PROVIDER="gemini"
-          DEFAULT_CHAT_MODEL="Gemini Flash 2.5"
-          DEFAULT_FAST_CHAT_PROVIDER="gemini"
-          DEFAULT_FAST_CHAT_MODEL="Gemini Flash 2.5"
-          DEFAULT_EMBEDDING_PROVIDER="openai"
-          DEFAULT_EMBEDDING_MODEL="Text embedding 3 large"
 
           # API Keys
           OPENAI_API_KEY="${{ secrets.OPENAI }}"
           ANTHROPIC_API_KEY="${{ secrets.ANTHROPIC }}"
           GEMINI_API_KEY="${{ secrets.GEMINI }}"
 
-          # Version Configuration
-          STARKNET_FOUNDRY_VERSION="0.47.0"
-          SCARB_VERSION="2.11.4"
           EOL
 
       - name: Generate embeddings

README.md

@@ -66,9 +66,10 @@ Using Docker is highly recommended for a streamlined setup.
 
     Edit the `.env` file with your credentials:
 
-    - Database credentials (defaults provided for local development)
-    - LLM API keys (at least one required: OpenAI, Anthropic, or Gemini)
-    - Optional: LangSmith credentials for monitoring
+    - Database credentials (defaults provided above for local development)
+    - LLM API keys (at least one required: `GEMINI_API_KEY` is recommended)
+    - Optional: `XAI_API_KEY` for Grok search functionality
+    - Optional: `LANGSMITH_*` variables for monitoring
 
 3.  **Run the Ingester (First Time Setup)**
 
@@ -120,12 +121,11 @@ Cairo Coder uses a modern architecture based on Retrieval-Augmented Generation (
 
 ### Project Structure
 
-The project is organized as a monorepo with multiple packages:
+The project is organized as a monorepo with multiple components:
 
 - **python/**: The core RAG agent and API server implementation using DSPy and FastAPI.
-- **packages/ingester/**: (TypeScript) Data ingestion tools for Cairo documentation sources.
-- **packages/typescript-config/**: Shared TypeScript configuration.
-- **(Legacy)** `packages/agents`: The original Langchain-based TypeScript implementation.
+- **ingesters/**: (TypeScript/Bun) Data ingestion tools for Cairo documentation sources.
+- **docker-compose.yml**: Orchestrates postgres, backend, and ingester services.
 
 ### RAG Pipeline (Python/DSPy)
 
@@ -141,21 +141,48 @@ The RAG pipeline is implemented in the `python/src/cairo_coder/core/` directory
 
 ### Python Service
 
-For local development of the Python service, navigate to `python/` and run the following commands`
+For local development of the Python service:
 
 1.  **Setup Environment**:
+
+    ```bash
+    cd python
+
+    # Install uv package manager (if not already installed)
+    curl -LsSf https://astral.sh/uv/install.sh | sh
+
+    # Install dependencies
+    uv sync
+    ```
+
+2.  **Start Database** (from root directory):
+
     ```bash
-    # Install uv package manager
-      curl -LsSf https://astral.sh/uv/install.sh | sh
+    # From root directory
+    docker compose up postgres
     ```
-2.  **Run Server**:
-    > Note: make sure the database is running, and the ingesters have been run.
+
+3.  **Run Ingester** (first time setup, from root directory):
+
     ```bash
-      uv run cairo-coder --dev
+    # From root directory
+    cd ingesters
+    bun install
+    bun run generate-embeddings:yes
+    cd ..
     ```
-3.  **Run Tests & Linting**:
+
+4.  **Run Server** (from python directory):
+
+    ```bash
+    cd python
+    uv run cairo-coder --dev
+    ```
+
+5.  **Run Tests** (from python directory):
     ```bash
-      uv run pytest
+    cd python
+    uv run pytest
     ```
 
 ### Starklings Evaluation
@@ -165,11 +192,12 @@ A script is included to evaluate the agent's performance on the Starklings exerc
 > Note: we recommend pre-warming the compilation cache by running `cd fixtures/runner_crate && scarb build` before running the evaluation.
 
 ```bash
-# Run a single evaluation round
+# From python directory
+cd python
 uv run starklings_evaluate
 ```
 
-Results are saved in the `starklings_results/` directory.
+Results are saved in the `python/starklings_results/` directory.
 
 ## Contribution
 

python/README.md

@@ -18,7 +18,9 @@ Cairo Coder is an AI-powered code generation service specifically designed for t
 ## Installation
 
 ```bash
-# Install uv package manager
+cd python
+
+# Install uv package manager (if not already installed)
 curl -LsSf https://astral.sh/uv/install.sh | sh
 
 # Install dependencies
@@ -31,8 +33,8 @@ Cairo Coder uses environment variables for all sensitive configuration like API
 
 ```bash
 # From the root directory
-cp .env.example .env
-# Edit .env with your credentials
+# Create .env file with required environment variables
+# See main README.md for required variables
 ```
 
 ## Running the Service
@@ -56,15 +58,26 @@ docker compose up postgres
 
 ```bash
 # From root directory
-pnpm generate-embeddings
+cd ingesters
+bun install
+bun run generate-embeddings:yes
+cd ..
 ```
 
-4. Start the FastAPI server:
+4. Start the FastAPI server (from python directory):
 
 ```bash
+cd python
 uv run cairo-coder
 ```
 
+Or with development mode (auto-reload):
+
+```bash
+cd python
+uv run cairo-coder --dev
+```
+
 ### Dockerized
 
 All configuration is handled automatically via Docker Compose. From the root directory:

python/src/cairo_coder/config/manager.py

@@ -27,7 +27,6 @@ def load_config() -> Config:
             user=os.getenv("POSTGRES_USER", "cairocoder"),
             password=os.getenv("POSTGRES_PASSWORD", ""),
             table_name=os.getenv("POSTGRES_TABLE_NAME", "documents"),
-            similarity_measure=os.getenv("SIMILARITY_MEASURE", "cosine"),
         )
 
         # Validate essential configuration

python/src/cairo_coder/core/config.py

@@ -13,8 +13,6 @@ class VectorStoreConfig:
     user: str
     password: str
     table_name: str
-    embedding_dimension: int = 2048  # text-embedding-3-large dimension
-    similarity_measure: str = "cosine"  # cosine, dot_product, euclidean
 
     @property
     def dsn(self) -> str:

python/src/cairo_coder/dspy/pgvector_rm.py

@@ -44,8 +44,8 @@ class PgVectorRM(dspy.Retrieve):
         llm = dspy.LM("gemini/gemini-flash-latest")
         dspy.configure(lm=llm)
 
-        # DATABASE_URL should be in the format postgresql://user:password@host/database
-        db_url = os.getenv("DATABASE_URL")
+        # DATABASE_URL should be in the format:
+        db_url = postgresql://user:password@host/database
 
         # embedding_func will default to dspy.settings.embedder
         retriever_model = PgVectorRM(db_url, "paragraphs", fields=["text", "document_id"], k=20)

python/src/cairo_coder/server/app.py

@@ -755,7 +755,6 @@ def get_vector_store_config() -> VectorStoreConfig:
         user=config.vector_store.user,
         password=config.vector_store.password,
         table_name=config.vector_store.table_name,
-        similarity_measure=config.vector_store.similarity_measure,
     )
 
 

python/tests/conftest.py

@@ -367,8 +367,6 @@ def clean_config_env_vars(monkeypatch):
         "POSTGRES_USER",
         "POSTGRES_PASSWORD",
         "POSTGRES_TABLE_NAME",
-        "POSTGRES_ROOT_DB",
-        "SIMILARITY_MEASURE",
         "HOST",
         "PORT",
         "DEBUG",

python/tests/integration/test_config_integration.py

@@ -17,7 +17,6 @@ def test_load_configuration_from_env(self, monkeypatch: pytest.MonkeyPatch) -> N
         monkeypatch.setenv("POSTGRES_USER", "test_user")
         monkeypatch.setenv("POSTGRES_PASSWORD", "test_password")
         monkeypatch.setenv("POSTGRES_TABLE_NAME", "test_documents")
-        monkeypatch.setenv("SIMILARITY_MEASURE", "cosine")
         monkeypatch.setenv("HOST", "localhost")
         monkeypatch.setenv("PORT", "8001")
         monkeypatch.setenv("DEBUG", "true")
@@ -31,7 +30,6 @@ def test_load_configuration_from_env(self, monkeypatch: pytest.MonkeyPatch) -> N
         assert config.vector_store.user == "test_user"
         assert config.vector_store.password == "test_password"
         assert config.vector_store.table_name == "test_documents"
-        assert config.vector_store.similarity_measure == "cosine"
         assert config.host == "localhost"
         assert config.port == 8001
         assert config.debug is True

python/tests/unit/test_config.py

@@ -30,7 +30,6 @@ def test_load_config_with_defaults(self, monkeypatch: pytest.MonkeyPatch) -> Non
         assert config.vector_store.user == "cairocoder"
         assert config.vector_store.password == "test-password"
         assert config.vector_store.table_name == "documents"
-        assert config.vector_store.similarity_measure == "cosine"
         assert config.host == "0.0.0.0"
         assert config.port == 3001
         assert config.debug is False
@@ -44,7 +43,6 @@ def test_load_config_from_environment(self, monkeypatch: pytest.MonkeyPatch) ->
         monkeypatch.setenv("POSTGRES_USER", "env-user")
         monkeypatch.setenv("POSTGRES_PASSWORD", "env-pass")
         monkeypatch.setenv("POSTGRES_TABLE_NAME", "env-table")
-        monkeypatch.setenv("SIMILARITY_MEASURE", "dot_product")
         monkeypatch.setenv("HOST", "127.0.0.1")
         monkeypatch.setenv("PORT", "8080")
         monkeypatch.setenv("DEBUG", "true")
@@ -58,7 +56,6 @@ def test_load_config_from_environment(self, monkeypatch: pytest.MonkeyPatch) ->
         assert config.vector_store.user == "env-user"
         assert config.vector_store.password == "env-pass"
         assert config.vector_store.table_name == "env-table"
-        assert config.vector_store.similarity_measure == "dot_product"
         assert config.host == "127.0.0.1"
         assert config.port == 8080
         assert config.debug is True