haiku.rag 0.12.0__tar.gz → 0.13.0__tar.gz

haiku_rag-0.13.0/.dockerignore

@@ -0,0 +1,66 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments (uv best practice)
+.venv/
+venv/
+env/
+
+# Node.js
+node_modules/
+.next/
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+
+# Data
+*.lancedb/
+data/
+
+# Docs
+mkdocs.yml
+docs/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Git
+.git/
+.gitignore
+
+# Development
+tests/
+.pytest_cache/
+.coverage
+htmlcov/
+src/evaluations/
+server.json
+# Examples
+examples/

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/.gitignore

@@ -16,8 +16,9 @@ tests/data/
 .pytest_cache/
 .ruff_cache/
 
-# environment variables
+# environment variables and config files
 .env
+haiku.rag.yaml
 TODO.md
 PLAN.md
 DEVNOTES.md
@@ -25,3 +26,6 @@ DEVNOTES.md
 # mcp registry
 .mcpregistry_github_token
 .mcpregistry_registry_token
+
+# MkDocs site directory when doing local docs builds
+site/

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: haiku.rag
-Version: 0.12.0
+Version: 0.13.0
 Summary: Agentic Retrieval Augmented Generation (RAG) with LanceDB
 Author-email: Yiorgis Gozadinos <ggozadinos@gmail.com>
 License: MIT
@@ -13,9 +13,8 @@ Classifier: Operating System :: MacOS
 Classifier: Operating System :: Microsoft :: Windows :: Windows 10
 Classifier: Operating System :: Microsoft :: Windows :: Windows 11
 Classifier: Operating System :: POSIX :: Linux
-Classifier: Programming Language :: Python :: 3.10
-Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Typing :: Typed
 Requires-Python: >=3.12
 Requires-Dist: docling>=2.56.1
@@ -24,8 +23,9 @@ Requires-Dist: httpx>=0.28.1
 Requires-Dist: lancedb>=0.25.2
 Requires-Dist: pydantic-ai>=1.0.18
 Requires-Dist: pydantic-graph>=1.0.18
-Requires-Dist: pydantic>=2.12.1
+Requires-Dist: pydantic>=2.12.2
 Requires-Dist: python-dotenv>=1.1.1
+Requires-Dist: pyyaml>=6.0.1
 Requires-Dist: rich>=14.2.0
 Requires-Dist: tiktoken>=0.12.0
 Requires-Dist: typer>=0.19.2
@@ -44,7 +44,7 @@ Retrieval-Augmented Generation (RAG) library built on LanceDB.
 
 `haiku.rag` is a Retrieval-Augmented Generation (RAG) library built to work with LanceDB as a local vector database. It uses LanceDB for storing embeddings and performs semantic (vector) search as well as full-text search combined through native hybrid search with Reciprocal Rank Fusion. Both open-source (Ollama) as well as commercial (OpenAI, VoyageAI) embedding providers are supported.
 
-> **Note**: Starting with version 0.7.0, haiku.rag uses LanceDB instead of SQLite. If you have an existing SQLite database, use `haiku-rag migrate old_database.sqlite` to migrate your data safely.
+> **Note**: Configuration now uses YAML files instead of environment variables. If you're upgrading from an older version, run `haiku-rag init-config --from-env` to migrate your `.env` file to `haiku.rag.yaml`. See [Configuration](https://ggozad.github.io/haiku.rag/configuration/) for details.
 
 ## Features
 
@@ -65,6 +65,7 @@ Retrieval-Augmented Generation (RAG) library built on LanceDB.
 
 ```bash
 # Install
+# Python 3.12 or newer required
 uv pip install haiku.rag
 
 # Add documents
@@ -98,14 +99,12 @@ haiku-rag research \
 # Rebuild database (re-chunk and re-embed all documents)
 haiku-rag rebuild
 
-# Migrate from SQLite to LanceDB
-haiku-rag migrate old_database.sqlite
-
 # Start server with file monitoring
-export MONITOR_DIRECTORIES="/path/to/docs"
-haiku-rag serve
+haiku-rag serve --monitor
 ```
 
+To customize settings, create a `haiku.rag.yaml` config file (see [Configuration](https://ggozad.github.io/haiku.rag/configuration/)).
+
 ## Python Usage
 
 ```python
@@ -197,18 +196,29 @@ haiku-rag a2aclient
 ```
 
 The A2A agent provides:
+
 - Multi-turn dialogue with context
 - Intelligent multi-search for complex questions
 - Source citations with titles and URIs
 - Full document retrieval on request
 
+## Examples
+
+See the [examples directory](examples/) for working examples:
+
+- **[Interactive Research Assistant](examples/ag-ui-research/)** - Full-stack research assistant with Pydantic AI and AG-UI featuring human-in-the-loop approval and real-time state synchronization
+- **[Docker Setup](examples/docker/)** - Complete Docker deployment with file monitoring, MCP server, and A2A agent
+- **[A2A Security](examples/a2a-security/)** - Authentication examples (API key, OAuth2, GitHub)
+
 ## Documentation
 
 Full documentation at: https://ggozad.github.io/haiku.rag/
 
 - [Installation](https://ggozad.github.io/haiku.rag/installation/) - Provider setup
-- [Configuration](https://ggozad.github.io/haiku.rag/configuration/) - Environment variables
+- [Configuration](https://ggozad.github.io/haiku.rag/configuration/) - YAML configuration
 - [CLI](https://ggozad.github.io/haiku.rag/cli/) - Command reference
 - [Python API](https://ggozad.github.io/haiku.rag/python/) - Complete API docs
 - [Agents](https://ggozad.github.io/haiku.rag/agents/) - QA agent and multi-agent research
+- [MCP Server](https://ggozad.github.io/haiku.rag/mcp/) - Model Context Protocol integration
+- [A2A Agent](https://ggozad.github.io/haiku.rag/a2a/) - Agent-to-Agent protocol support
 - [Benchmarks](https://ggozad.github.io/haiku.rag/benchmarks/) - Performance Benchmarks

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/README.md

@@ -4,7 +4,7 @@ Retrieval-Augmented Generation (RAG) library built on LanceDB.
 
 `haiku.rag` is a Retrieval-Augmented Generation (RAG) library built to work with LanceDB as a local vector database. It uses LanceDB for storing embeddings and performs semantic (vector) search as well as full-text search combined through native hybrid search with Reciprocal Rank Fusion. Both open-source (Ollama) as well as commercial (OpenAI, VoyageAI) embedding providers are supported.
 
-> **Note**: Starting with version 0.7.0, haiku.rag uses LanceDB instead of SQLite. If you have an existing SQLite database, use `haiku-rag migrate old_database.sqlite` to migrate your data safely.
+> **Note**: Configuration now uses YAML files instead of environment variables. If you're upgrading from an older version, run `haiku-rag init-config --from-env` to migrate your `.env` file to `haiku.rag.yaml`. See [Configuration](https://ggozad.github.io/haiku.rag/configuration/) for details.
 
 ## Features
 
@@ -25,6 +25,7 @@ Retrieval-Augmented Generation (RAG) library built on LanceDB.
 
 ```bash
 # Install
+# Python 3.12 or newer required
 uv pip install haiku.rag
 
 # Add documents
@@ -58,14 +59,12 @@ haiku-rag research \
 # Rebuild database (re-chunk and re-embed all documents)
 haiku-rag rebuild
 
-# Migrate from SQLite to LanceDB
-haiku-rag migrate old_database.sqlite
-
 # Start server with file monitoring
-export MONITOR_DIRECTORIES="/path/to/docs"
-haiku-rag serve
+haiku-rag serve --monitor
 ```
 
+To customize settings, create a `haiku.rag.yaml` config file (see [Configuration](https://ggozad.github.io/haiku.rag/configuration/)).
+
 ## Python Usage
 
 ```python
@@ -157,18 +156,29 @@ haiku-rag a2aclient
 ```
 
 The A2A agent provides:
+
 - Multi-turn dialogue with context
 - Intelligent multi-search for complex questions
 - Source citations with titles and URIs
 - Full document retrieval on request
 
+## Examples
+
+See the [examples directory](examples/) for working examples:
+
+- **[Interactive Research Assistant](examples/ag-ui-research/)** - Full-stack research assistant with Pydantic AI and AG-UI featuring human-in-the-loop approval and real-time state synchronization
+- **[Docker Setup](examples/docker/)** - Complete Docker deployment with file monitoring, MCP server, and A2A agent
+- **[A2A Security](examples/a2a-security/)** - Authentication examples (API key, OAuth2, GitHub)
+
 ## Documentation
 
 Full documentation at: https://ggozad.github.io/haiku.rag/
 
 - [Installation](https://ggozad.github.io/haiku.rag/installation/) - Provider setup
-- [Configuration](https://ggozad.github.io/haiku.rag/configuration/) - Environment variables
+- [Configuration](https://ggozad.github.io/haiku.rag/configuration/) - YAML configuration
 - [CLI](https://ggozad.github.io/haiku.rag/cli/) - Command reference
 - [Python API](https://ggozad.github.io/haiku.rag/python/) - Complete API docs
 - [Agents](https://ggozad.github.io/haiku.rag/agents/) - QA agent and multi-agent research
+- [MCP Server](https://ggozad.github.io/haiku.rag/mcp/) - Model Context Protocol integration
+- [A2A Agent](https://ggozad.github.io/haiku.rag/a2a/) - Agent-to-Agent protocol support
 - [Benchmarks](https://ggozad.github.io/haiku.rag/benchmarks/) - Performance Benchmarks

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/mkdocs.yml

@@ -57,6 +57,7 @@ plugins:
 nav:
   - haiku.rag:
       - index.md
+      - Getting started: tutorial.md
       - Installation: installation.md
       - Configuration: configuration.md
       - CLI: cli.md

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/pyproject.toml

@@ -2,7 +2,7 @@
 
 name = "haiku.rag"
 description = "Agentic Retrieval Augmented Generation (RAG) with LanceDB"
-version = "0.12.0"
+version = "0.13.0"
 authors = [{ name = "Yiorgis Gozadinos", email = "ggozadinos@gmail.com" }]
 license = { text = "MIT" }
 readme = { file = "README.md", content-type = "text/markdown" }
@@ -16,9 +16,8 @@ classifiers = [
     "Operating System :: Microsoft :: Windows :: Windows 11",
     "Operating System :: MacOS",
     "Operating System :: POSIX :: Linux",
-    "Programming Language :: Python :: 3.10",
-    "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
     "Typing :: Typed",
 ]
 
@@ -27,10 +26,11 @@ dependencies = [
     "fastmcp>=2.12.4",
     "httpx>=0.28.1",
     "lancedb>=0.25.2",
-    "pydantic>=2.12.1",
+    "pydantic>=2.12.2",
     "pydantic-ai>=1.0.18",
     "pydantic-graph>=1.0.18",
     "python-dotenv>=1.1.1",
+    "pyyaml>=6.0.1",
     "rich>=14.2.0",
     "tiktoken>=0.12.0",
     "typer>=0.19.2",
@@ -50,7 +50,7 @@ requires = ["hatchling"]
 build-backend = "hatchling.build"
 
 [tool.hatch.build]
-exclude = ["/docs", "/examples", "/tests", "/.github"]
+exclude = ["/docs", "/examples", "/tests", "/docker", "/.github"]
 
 [tool.hatch.build.targets.wheel]
 packages = ["src/haiku"]
@@ -63,7 +63,7 @@ dev = [
     "mkdocs-material>=9.6.14",
     "pydantic-evals>=1.0.8",
     "pre-commit>=4.2.0",
-    "pyright>=1.1.405",
+    "pyright>=1.1.406",
     "pytest>=8.4.2",
     "pytest-asyncio>=1.2.0",
     "pytest-cov>=7.0.0",

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/src/evaluations/benchmark.py

@@ -174,7 +174,7 @@ async def run_qa_benchmark(
 
     judge_model = OpenAIChatModel(
         model_name=QA_JUDGE_MODEL,
-        provider=OllamaProvider(base_url=f"{Config.OLLAMA_BASE_URL}/v1"),
+        provider=OllamaProvider(base_url=f"{Config.providers.ollama.base_url}/v1"),
     )
 
     evaluation_dataset = EvalDataset[str, str, dict[str, str]](

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/src/evaluations/llm_judge.py

@@ -41,7 +41,7 @@ class LLMJudge:
         # Create Ollama model
         ollama_model = OpenAIChatModel(
             model_name=model,
-            provider=OllamaProvider(base_url=f"{Config.OLLAMA_BASE_URL}/v1"),
+            provider=OllamaProvider(base_url=f"{Config.providers.ollama.base_url}/v1"),
         )
 
         # Create Pydantic AI agent

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/src/haiku/rag/a2a/__init__.py

@@ -57,12 +57,12 @@ def create_a2a_app(
     """
     base_storage = InMemoryStorage()
     storage = LRUMemoryStorage(
-        storage=base_storage, max_contexts=Config.A2A_MAX_CONTEXTS
+        storage=base_storage, max_contexts=Config.a2a.max_contexts
     )
     broker = InMemoryBroker()
 
     # Create the agent with native search tool
-    model = get_model(Config.QA_PROVIDER, Config.QA_MODEL)
+    model = get_model(Config.qa.provider, Config.qa.model)
     agent = Agent(
         model=model,
         deps_type=AgentDependencies,
@@ -120,7 +120,7 @@ def create_a2a_app(
     # Create FastA2A app with custom worker lifecycle
     @asynccontextmanager
     async def lifespan(app):
-        logger.info(f"Started A2A server (max contexts: {Config.A2A_MAX_CONTEXTS})")
+        logger.info(f"Started A2A server (max contexts: {Config.a2a.max_contexts})")
         async with app.task_manager:
             async with worker.run():
                 yield

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/src/haiku/rag/a2a/client.py

@@ -7,9 +7,18 @@ from rich.console import Console
 from rich.markdown import Markdown
 from rich.prompt import Prompt
 
+try:
+    from fasta2a.client import A2AClient as FastA2AClient
+    from fasta2a.schema import Message, TextPart
+except ImportError as e:
+    raise ImportError(
+        "A2A support requires the 'a2a' extra. "
+        "Install with: uv pip install 'haiku.rag[a2a]'"
+    ) from e
+
 
 class A2AClient:
-    """Simple A2A protocol client."""
+    """Interactive A2A protocol client."""
 
     def __init__(self, base_url: str = "http://localhost:8000"):
         """Initialize A2A client.
@@ -18,11 +27,12 @@ class A2AClient:
             base_url: Base URL of the A2A server
         """
         self.base_url = base_url.rstrip("/")
-        self.client = httpx.AsyncClient(timeout=60.0)
+        http_client = httpx.AsyncClient(timeout=60.0)
+        self._client = FastA2AClient(base_url=base_url, http_client=http_client)
 
     async def close(self):
         """Close the HTTP client."""
-        await self.client.aclose()
+        await self._client.http_client.aclose()
 
     async def get_agent_card(self) -> dict[str, Any]:
         """Fetch the agent card from the A2A server.
@@ -30,7 +40,9 @@ class A2AClient:
         Returns:
             Agent card dictionary with agent capabilities and metadata
         """
-        response = await self.client.get(f"{self.base_url}/.well-known/agent-card.json")
+        response = await self._client.http_client.get(
+            f"{self.base_url}/.well-known/agent-card.json"
+        )
         response.raise_for_status()
         return response.json()
 
@@ -53,46 +65,38 @@ class A2AClient:
         if context_id is None:
             context_id = str(uuid.uuid4())
 
-        message_id = str(uuid.uuid4())
-
-        payload: dict[str, Any] = {
-            "jsonrpc": "2.0",
-            "method": "message/send",
-            "params": {
-                "contextId": context_id,
-                "message": {
-                    "kind": "message",
-                    "role": "user",
-                    "messageId": message_id,
-                    "parts": [{"kind": "text", "text": text}],
-                },
-            },
-            "id": 1,
-        }
+        message = Message(
+            kind="message",
+            role="user",
+            message_id=str(uuid.uuid4()),
+            parts=[TextPart(kind="text", text=text)],
+        )
 
+        metadata: dict[str, Any] = {"contextId": context_id}
         if skill_id:
-            payload["params"]["skillId"] = skill_id
+            metadata["skillId"] = skill_id
 
-        response = await self.client.post(
-            self.base_url,
-            json=payload,
-            headers={"Content-Type": "application/json"},
-        )
-        response.raise_for_status()
-        initial_response = response.json()
+        response = await self._client.send_message(message, metadata=metadata)
 
-        # Extract task ID from response
-        result = initial_response.get("result", {})
-        task_id = result.get("id")
+        if "error" in response:
+            return {"error": response["error"]}
 
-        if not task_id:
-            return initial_response
+        result = response.get("result")
+        if not result:
+            return {"result": result}
 
-        # Poll for task completion
-        return await self.wait_for_task(task_id)
+        # Result can be either Task or Message - check if it's a Task with an id
+        if result.get("kind") == "task":
+            task_id = result.get("id")
+            if task_id:
+                # Poll for task completion
+                return await self.wait_for_task(task_id)
+
+        # Return the message directly
+        return {"result": result}
 
     async def wait_for_task(
-        self, task_id: str, max_wait: int = 60, poll_interval: float = 0.5
+        self, task_id: str, max_wait: int = 120, poll_interval: float = 0.5
     ) -> dict[str, Any]:
         """Poll for task completion.
 
@@ -109,27 +113,19 @@ class A2AClient:
         start_time = time.time()
 
         while time.time() - start_time < max_wait:
-            payload = {
-                "jsonrpc": "2.0",
-                "method": "tasks/get",
-                "params": {"id": task_id},
-                "id": 2,
-            }
-
-            response = await self.client.post(
-                self.base_url,
-                json=payload,
-                headers={"Content-Type": "application/json"},
-            )
-            response.raise_for_status()
-            task = response.json()
-
-            result = task.get("result", {})
-            status = result.get("status", {})
-            state = status.get("state")
+            task_response = await self._client.get_task(task_id)
+
+            if "error" in task_response:
+                return {"error": task_response["error"]}
+
+            task = task_response.get("result")
+            if not task:
+                raise Exception("No task in response")
+
+            state = task.get("status", {}).get("state")
 
             if state == "completed":
-                return task
+                return {"result": task}
             elif state == "failed":
                 raise Exception(f"Task failed: {task}")
 
@@ -191,6 +187,7 @@ def print_response(response: dict[str, Any], console: Console):
 
     # Print artifacts summary with details
     if artifacts:
+        console.rule("[dim]Artifacts generated[/dim]")
         summary_lines = []
 
         for artifact in artifacts:

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/src/haiku/rag/app.py

@@ -160,13 +160,20 @@ class HaikuRAGApp:
         self, source: str, title: str | None = None, metadata: dict | None = None
     ):
         async with HaikuRAG(db_path=self.db_path) as self.client:
-            doc = await self.client.create_document_from_source(
+            result = await self.client.create_document_from_source(
                 source, title=title, metadata=metadata
             )
-            self._rich_print_document(doc, truncate=True)
-            self.console.print(
-                f"[bold green]Document {doc.id} added successfully.[/bold green]"
-            )
+            if isinstance(result, list):
+                for doc in result:
+                    self._rich_print_document(doc, truncate=True)
+                self.console.print(
+                    f"[bold green]{len(result)} documents added successfully.[/bold green]"
+                )
+            else:
+                self._rich_print_document(result, truncate=True)
+                self.console.print(
+                    f"[bold green]Document {result.id} added successfully.[/bold green]"
+                )
 
     async def get_document(self, doc_id: str):
         async with HaikuRAG(db_path=self.db_path) as self.client:
@@ -224,8 +231,8 @@ class HaikuRAGApp:
                     )
 
                     start_node = DeepQAPlanNode(
-                        provider=Config.QA_PROVIDER,
-                        model=Config.QA_MODEL,
+                        provider=Config.qa.provider,
+                        model=Config.qa.model,
                     )
 
                     result = await graph.run(
@@ -271,8 +278,8 @@ class HaikuRAGApp:
                 )
 
                 start = PlanNode(
-                    provider=Config.RESEARCH_PROVIDER or Config.QA_PROVIDER,
-                    model=Config.RESEARCH_MODEL or Config.QA_MODEL,
+                    provider=Config.research.provider or Config.qa.provider,
+                    model=Config.research.model or Config.qa.model,
                 )
                 report = None
                 async for event in stream_research_graph(graph, start, state, deps):
@@ -467,7 +474,9 @@ class HaikuRAGApp:
 
             # Start file monitor if enabled
             if enable_monitor:
-                monitor = FileWatcher(paths=Config.MONITOR_DIRECTORIES, client=client)
+                monitor = FileWatcher(
+                    paths=Config.storage.monitor_directories, client=client
+                )
                 monitor_task = asyncio.create_task(monitor.observe())
                 tasks.append(monitor_task)
 

{haiku_rag-0.12.0 → haiku_rag-0.13.0}/src/haiku/rag/chunker.py

@@ -22,7 +22,7 @@ class Chunker:
 
     def __init__(
         self,
-        chunk_size: int = Config.CHUNK_SIZE,
+        chunk_size: int = Config.processing.chunk_size,
     ):
         self.chunk_size = chunk_size
         tokenizer = OpenAITokenizer(