haiku.rag 0.9.1__tar.gz → 0.9.2__tar.gz

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: haiku.rag
-Version: 0.9.1
+Version: 0.9.2
 Summary: Agentic Retrieval Augmented Generation (RAG) with LanceDB
 Author-email: Yiorgis Gozadinos <ggozadinos@gmail.com>
 License: MIT

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/docs/agents.md

@@ -43,6 +43,8 @@ The research workflow coordinates specialized agents to plan, search, analyze, a
 Components:
 
 - Orchestrator: Plans, coordinates, and loops until confidence is sufficient
+- Presearch Survey: Runs a quick KB scan and summarizes relevant chunk text to
+  ground the initial plan (plain-text summary; no URIs or scores)
 - Search Specialist: Performs targeted RAG searches and answers sub‑questions
 - Analysis & Evaluation: Extracts insights, identifies gaps, proposes new questions
 - Synthesis: Produces a final structured research report

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "haiku.rag"
-version = "0.9.1"
+version = "0.9.2"
 description = "Agentic Retrieval Augmented Generation (RAG) with LanceDB"
 authors = [{ name = "Yiorgis Gozadinos", email = "ggozadinos@gmail.com" }]
 license = { text = "MIT" }

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/app.py

@@ -122,12 +122,7 @@ class HaikuRAGApp:
                         self.console.print(f"• {finding}")
                     self.console.print()
 
-                # Themes
-                if report.themes:
-                    self.console.print("[bold cyan]Key Themes:[/bold cyan]")
-                    for theme, explanation in report.themes.items():
-                        self.console.print(f"• [bold]{theme}[/bold]: {explanation}")
-                    self.console.print()
+                # (Themes section removed)
 
                 # Conclusions
                 if report.conclusions:
@@ -261,7 +256,7 @@ class HaikuRAGApp:
                 elif transport == "sse":
                     await server.run_sse_async()
                 else:
-                    await server.run_http_async("streamable-http")
+                    await server.run_http_async(transport="streamable-http")
             except KeyboardInterrupt:
                 pass
             finally:

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/qa/agent.py

@@ -49,6 +49,9 @@ class QuestionAnswerAgent:
             limit: int = 3,
         ) -> list[SearchResult]:
             """Search the knowledge base for relevant documents."""
+
+            # Remove quotes from queries as this requires positional indexing in lancedb
+            query = query.replace('"', "")
             search_results = await ctx.deps.client.search(query, limit=limit)
             expanded_results = await ctx.deps.client.expand_context(search_results)
 

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/__init__.py

@@ -12,6 +12,7 @@ from haiku.rag.research.evaluation_agent import (
     EvaluationResult,
 )
 from haiku.rag.research.orchestrator import ResearchOrchestrator, ResearchPlan
+from haiku.rag.research.presearch_agent import PresearchSurveyAgent
 from haiku.rag.research.search_agent import SearchSpecialistAgent
 from haiku.rag.research.synthesis_agent import ResearchReport, SynthesisAgent
 
@@ -25,6 +26,7 @@ __all__ = [
     # Specialized agents
     "SearchAnswer",
     "SearchSpecialistAgent",
+    "PresearchSurveyAgent",
     "AnalysisEvaluationAgent",
     "EvaluationResult",
     "SynthesisAgent",

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/base.py

@@ -33,10 +33,18 @@ class BaseResearchAgent[T](ABC):
         # Import deps type lazily to avoid circular import during module load
         from haiku.rag.research.dependencies import ResearchDependencies
 
+        # If the agent is expected to return plain text, pass `str` directly.
+        # Otherwise, wrap the model with ToolOutput for robust tool-handling retries.
+        agent_output_type: Any
+        if self.output_type is str:  # plain text output
+            agent_output_type = str
+        else:
+            agent_output_type = ToolOutput(self.output_type, max_retries=3)
+
         self._agent = Agent(
             model=model_obj,
             deps_type=ResearchDependencies,
-            output_type=ToolOutput(self.output_type, max_retries=3),
+            output_type=agent_output_type,
             system_prompt=self.get_system_prompt(),
         )
 

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/evaluation_agent.py

@@ -11,7 +11,9 @@ class EvaluationResult(BaseModel):
         description="Main insights extracted from the research so far"
     )
     new_questions: list[str] = Field(
-        description="New sub-questions to add to the research (max 3)", max_length=3
+        description="New sub-questions to add to the research (max 3)",
+        max_length=3,
+        default=[],
     )
     confidence_score: float = Field(
         description="Confidence level in the completeness of research (0-1)",

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/orchestrator.py

@@ -12,6 +12,7 @@ from haiku.rag.research.evaluation_agent import (
     AnalysisEvaluationAgent,
     EvaluationResult,
 )
+from haiku.rag.research.presearch_agent import PresearchSurveyAgent
 from haiku.rag.research.prompts import ORCHESTRATOR_PROMPT
 from haiku.rag.research.search_agent import SearchSpecialistAgent
 from haiku.rag.research.synthesis_agent import ResearchReport, SynthesisAgent
@@ -41,6 +42,9 @@ class ResearchOrchestrator(BaseResearchAgent[ResearchPlan]):
         self.search_agent: SearchSpecialistAgent = SearchSpecialistAgent(
             provider, model
         )
+        self.presearch_agent: PresearchSurveyAgent = PresearchSurveyAgent(
+            provider, model
+        )
         self.evaluation_agent: AnalysisEvaluationAgent = AnalysisEvaluationAgent(
             provider, model
         )
@@ -61,7 +65,12 @@ class ResearchOrchestrator(BaseResearchAgent[ResearchPlan]):
             "original_question": context.original_question,
             "unanswered_questions": context.sub_questions,
             "qa_responses": [
-                {"question": qa.query, "answer": qa.answer}
+                {
+                    "question": qa.query,
+                    "answer": qa.answer,
+                    "context_snippets": qa.context,
+                    "sources": qa.sources,
+                }
                 for qa in context.qa_responses
             ],
             "insights": context.insights,
@@ -99,12 +108,38 @@ class ResearchOrchestrator(BaseResearchAgent[ResearchPlan]):
         # Use provided console or create a new one
         console = console or Console() if verbose else None
 
+        # Run a simple presearch survey to summarize KB context
+        if console:
+            console.print(
+                "\n[bold cyan]🔎 Presearch: summarizing KB context...[/bold cyan]"
+            )
+
+        presearch_result = await self.presearch_agent.run(question, deps=deps)
+
         # Create initial research plan
         if console:
             console.print("\n[bold cyan]📋 Creating research plan...[/bold cyan]")
 
+        # Include the presearch summary to ground the planning step.
+
+        planning_context_xml = format_as_xml(
+            {
+                "original_question": question,
+                "presearch_summary": presearch_result.output or "",
+            },
+            root_tag="planning_context",
+        )
+
+        plan_prompt = (
+            "Create a research plan for the main question below.\n\n"
+            f"Main question: {question}\n\n"
+            "Use this brief presearch summary to inform the plan. Focus the 3 sub-questions "
+            "on the most important aspects not already obvious from the current KB context.\n\n"
+            f"{planning_context_xml}"
+        )
+
         plan_result: AgentRunResult[ResearchPlan] = await self.run(
-            f"Create a research plan for: {question}", deps=deps
+            plan_prompt, deps=deps
         )
 
         context.sub_questions = plan_result.output.sub_questions

haiku_rag-0.9.2/src/haiku/rag/research/presearch_agent.py

@@ -0,0 +1,34 @@
+from pydantic_ai import RunContext
+from pydantic_ai.run import AgentRunResult
+
+from haiku.rag.research.base import BaseResearchAgent
+from haiku.rag.research.dependencies import ResearchDependencies
+from haiku.rag.research.prompts import PRESEARCH_AGENT_PROMPT
+
+
+class PresearchSurveyAgent(BaseResearchAgent[str]):
+    """Presearch agent that gathers verbatim context and summarizes it."""
+
+    def __init__(self, provider: str, model: str) -> None:
+        super().__init__(provider, model, str)
+
+    async def run(
+        self, prompt: str, deps: ResearchDependencies, **kwargs
+    ) -> AgentRunResult[str]:
+        return await super().run(prompt, deps, **kwargs)
+
+    def get_system_prompt(self) -> str:
+        return PRESEARCH_AGENT_PROMPT
+
+    def register_tools(self) -> None:
+        @self.agent.tool
+        async def gather_context(
+            ctx: RunContext[ResearchDependencies],
+            query: str,
+            limit: int = 6,
+        ) -> str:
+            """Return verbatim concatenation of relevant chunk texts."""
+            query = query.replace('"', "")
+            results = await ctx.deps.client.search(query, limit=limit)
+            expanded = await ctx.deps.client.expand_context(results)
+            return "\n\n".join(chunk.content for chunk, _ in expanded)

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/prompts.py

@@ -114,3 +114,16 @@ Focus on creating a report that provides clear value to the reader by:
 - Highlighting the most important findings
 - Explaining the implications of the research
 - Suggesting concrete next steps"""
+
+PRESEARCH_AGENT_PROMPT = """You are a rapid research surveyor.
+
+Task:
+- Call the gather_context tool once with the main question to obtain a
+  relevant texts from the Knowledge Base (KB).
+- Read that context and produce a brief natural-language summary describing
+  what the KB appears to contain relative to the question.
+
+Rules:
+- Base the summary strictly on the provided text; do not invent.
+- Output only the summary as plain text (one short paragraph).
+"""

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/search_agent.py

@@ -42,6 +42,7 @@ class SearchSpecialistAgent(BaseResearchAgent[SearchAnswer]):
         ) -> str:
             """Search the KB and return a concise context pack."""
             # Remove quotes from queries as this requires positional indexing in lancedb
+            # XXX: Investigate how to do that with lancedb
             query = query.replace('"', "")
             search_results = await ctx.deps.client.search(query, limit=limit)
             expanded = await ctx.deps.client.expand_context(search_results)

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/src/haiku/rag/research/synthesis_agent.py

@@ -12,11 +12,12 @@ class ResearchReport(BaseModel):
     main_findings: list[str] = Field(
         description="Primary research findings with supporting evidence"
     )
-    themes: dict[str, str] = Field(description="Major themes and their explanations")
     conclusions: list[str] = Field(description="Evidence-based conclusions")
-    limitations: list[str] = Field(description="Limitations of the current research")
+    limitations: list[str] = Field(
+        description="Limitations of the current research", default=[]
+    )
     recommendations: list[str] = Field(
-        description="Actionable recommendations based on findings"
+        description="Actionable recommendations based on findings", default=[]
     )
     sources_summary: str = Field(
         description="Summary of sources used and their reliability"

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/tests/research/test_orchestrator.py

@@ -172,7 +172,6 @@ class TestResearchOrchestrator:
                         assert report.title
                         assert report.executive_summary
                         assert isinstance(report.main_findings, list)
-                        assert isinstance(report.themes, dict)
                         assert isinstance(report.conclusions, list)
                         assert isinstance(report.limitations, list)
                         assert isinstance(report.recommendations, list)

{haiku_rag-0.9.1 → haiku_rag-0.9.2}/uv.lock

@@ -1111,7 +1111,7 @@ wheels = [
 
 [[package]]
 name = "haiku-rag"
-version = "0.9.1"
+version = "0.9.2"
 source = { editable = "." }
 dependencies = [
     { name = "docling" },