bridgekit 0.1.1__tar.gz → 0.2.0__tar.gz

{bridgekit-0.1.1/bridgekit.egg-info → bridgekit-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bridgekit
-Version: 0.1.1
+Version: 0.2.0
 Summary: AI tools that make you a better data scientist, not a redundant one.
 License: MIT
 Project-URL: Homepage, https://github.com/getbridgekit/bridgekit
@@ -15,6 +15,12 @@ Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: anthropic>=0.20.0
+Requires-Dist: chromadb>=0.4.0
+Requires-Dist: sentence-transformers>=2.0.0
+Requires-Dist: pypdf>=3.0.0
+Requires-Dist: python-docx>=1.0.0
+Requires-Dist: python-pptx>=0.6.0
+Requires-Dist: nbformat>=5.0.0
 Dynamic: license-file
 
 # Bridgekit
@@ -134,6 +140,47 @@ Paste your writeup as a string and call `evaluate()` — that's it.
 
 ---
 
+## Tool #2: Analysis Search
+
+Ask questions across a collection of your past analysis documents. Point it at a folder and get answers grounded in your actual work — no digging through files manually.
+
+Uses a vector database and semantic similarity to find relevant context across your documents — not keyword matching.
+
+Supports `.txt`, `.md`, `.pdf`, `.docx`, `.pptx`, and `.ipynb` files.
+
+**From a folder:**
+```python
+from bridgekit import ask
+
+print(ask("what drove churn in Q3?", source="reports/"))
+```
+
+**From raw text:**
+```python
+from bridgekit import ask
+
+text = """
+Q3 churn rose to 4.5%, driven by a product outage in August and a pricing
+change in July that increased SMB costs by 12%.
+"""
+
+print(ask("what caused the Q3 churn spike?", text=text))
+```
+
+**Output** *(based on sample data included in the repo)*:
+```
+Based on the Q3 2024 Churn Analysis, two primary factors drove the elevated
+churn rate of 4.5%:
+
+1. August Product Outage — A 14-hour outage affected 3,800 accounts. Impacted
+   accounts churned at 8.1% vs 3.2% for unaffected accounts.
+
+2. July Pricing Change — SMB costs increased by an average of 12%, causing SMB
+   churn to spike to 7.2% — the highest single-month figure in the dataset.
+```
+
+---
+
 ## Why not just use Claude?
 
 You could. But you'd need to know what to ask, how to frame it, and what a good answer looks like. Bridgekit has that baked in — it knows you're a data scientist presenting findings, so it asks the right questions automatically. No prompt engineering required. Just paste your work and run it.
@@ -156,7 +203,7 @@ Bridgekit only ever sees text you write yourself — your narrative, your conclu
 
 ## What's next?
 
-Bridgekit is a suite, not a one-off. The analysis reviewer is tool #1. Coming next:
+Bridgekit is a suite, not a one-off. Two tools are live — more are coming:
 
 - **Statistical approach suggester** — describe your problem in plain English, get the right test and why
 - **Stakeholder translator** — turn your technical findings into a narrative a non-technical audience will actually follow

bridgekit-0.1.1/PKG-INFO → bridgekit-0.2.0/README.md

@@ -1,22 +1,3 @@
-Metadata-Version: 2.4
-Name: bridgekit
-Version: 0.1.1
-Summary: AI tools that make you a better data scientist, not a redundant one.
-License: MIT
-Project-URL: Homepage, https://github.com/getbridgekit/bridgekit
-Project-URL: Issues, https://github.com/getbridgekit/bridgekit/issues
-Keywords: data science,AI,analysis,evaluation,anthropic
-Classifier: Development Status :: 3 - Alpha
-Classifier: Intended Audience :: Science/Research
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Classifier: Programming Language :: Python :: 3
-Classifier: License :: OSI Approved :: MIT License
-Requires-Python: >=3.9
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: anthropic>=0.20.0
-Dynamic: license-file
-
 # Bridgekit
 
 **AI tools that make you a better data scientist, not a redundant one.**
@@ -134,6 +115,47 @@ Paste your writeup as a string and call `evaluate()` — that's it.
 
 ---
 
+## Tool #2: Analysis Search
+
+Ask questions across a collection of your past analysis documents. Point it at a folder and get answers grounded in your actual work — no digging through files manually.
+
+Uses a vector database and semantic similarity to find relevant context across your documents — not keyword matching.
+
+Supports `.txt`, `.md`, `.pdf`, `.docx`, `.pptx`, and `.ipynb` files.
+
+**From a folder:**
+```python
+from bridgekit import ask
+
+print(ask("what drove churn in Q3?", source="reports/"))
+```
+
+**From raw text:**
+```python
+from bridgekit import ask
+
+text = """
+Q3 churn rose to 4.5%, driven by a product outage in August and a pricing
+change in July that increased SMB costs by 12%.
+"""
+
+print(ask("what caused the Q3 churn spike?", text=text))
+```
+
+**Output** *(based on sample data included in the repo)*:
+```
+Based on the Q3 2024 Churn Analysis, two primary factors drove the elevated
+churn rate of 4.5%:
+
+1. August Product Outage — A 14-hour outage affected 3,800 accounts. Impacted
+   accounts churned at 8.1% vs 3.2% for unaffected accounts.
+
+2. July Pricing Change — SMB costs increased by an average of 12%, causing SMB
+   churn to spike to 7.2% — the highest single-month figure in the dataset.
+```
+
+---
+
 ## Why not just use Claude?
 
 You could. But you'd need to know what to ask, how to frame it, and what a good answer looks like. Bridgekit has that baked in — it knows you're a data scientist presenting findings, so it asks the right questions automatically. No prompt engineering required. Just paste your work and run it.
@@ -156,7 +178,7 @@ Bridgekit only ever sees text you write yourself — your narrative, your conclu
 
 ## What's next?
 
-Bridgekit is a suite, not a one-off. The analysis reviewer is tool #1. Coming next:
+Bridgekit is a suite, not a one-off. Two tools are live — more are coming:
 
 - **Statistical approach suggester** — describe your problem in plain English, get the right test and why
 - **Stakeholder translator** — turn your technical findings into a narrative a non-technical audience will actually follow

bridgekit-0.2.0/bridgekit/__init__.py

@@ -0,0 +1,5 @@
+from .reviewer import evaluate
+from .search import ask
+
+__version__ = "0.2.0"
+__all__ = ["evaluate", "ask"]

bridgekit-0.2.0/bridgekit/search.py

@@ -0,0 +1,122 @@
+import os
+from pathlib import Path
+import anthropic
+import chromadb
+from chromadb.utils.embedding_functions import SentenceTransformerEmbeddingFunction
+
+CHUNK_SIZE = 150  # words per chunk
+CHUNK_OVERLAP = 20
+
+
+def _load_file(path: Path) -> str:
+    suffix = path.suffix.lower()
+    if suffix == ".pdf":
+        import pypdf
+        reader = pypdf.PdfReader(str(path))
+        return "\n".join(page.extract_text() or "" for page in reader.pages)
+    elif suffix == ".docx":
+        import docx
+        doc = docx.Document(str(path))
+        return "\n".join(p.text for p in doc.paragraphs if p.text.strip())
+    elif suffix == ".pptx":
+        from pptx import Presentation
+        prs = Presentation(str(path))
+        lines = []
+        for slide in prs.slides:
+            for shape in slide.shapes:
+                if hasattr(shape, "text") and shape.text.strip():
+                    lines.append(shape.text)
+        return "\n".join(lines)
+    elif suffix == ".ipynb":
+        import nbformat
+        nb = nbformat.read(str(path), as_version=4)
+        lines = []
+        for cell in nb.cells:
+            if cell.cell_type in ("markdown", "code") and cell.source.strip():
+                lines.append(cell.source)
+        return "\n\n".join(lines)
+    else:
+        return path.read_text(encoding="utf-8")
+
+
+def _chunk(text: str) -> list[str]:
+    words = text.split()
+    chunks = []
+    i = 0
+    while i < len(words):
+        chunks.append(" ".join(words[i:i + CHUNK_SIZE]))
+        i += CHUNK_SIZE - CHUNK_OVERLAP
+    return [c for c in chunks if c.strip()]
+
+
+def ask(question: str, source: str = None, text: str = None) -> str:
+    """
+    Ask a question across a collection of analysis documents or raw text.
+
+    Args:
+        question: The question to answer.
+        source:   Path to a folder containing .txt, .md, .pdf, .docx, .pptx, or .ipynb files.
+        text:     A raw text string to search instead of a folder.
+
+    Returns:
+        An answer grounded in the provided documents.
+    """
+    if not source and not text:
+        raise ValueError("Provide either 'source' (folder path) or 'text'.")
+
+    api_key = os.environ.get("ANTHROPIC_API_KEY")
+    if not api_key:
+        raise EnvironmentError(
+            "ANTHROPIC_API_KEY not found. Set it with: export ANTHROPIC_API_KEY=your_key_here"
+        )
+
+    # Collect chunks
+    chunks = []
+
+    if text:
+        chunks.extend(_chunk(text))
+
+    if source:
+        folder = Path(source).expanduser().resolve()
+        supported = {".txt", ".md", ".pdf", ".docx", ".pptx", ".ipynb"}
+        for file in sorted(folder.iterdir()):
+            if file.suffix.lower() in supported:
+                content = _load_file(file)
+                chunks.extend(_chunk(content))
+
+    if not chunks:
+        raise ValueError("No content found. Check your source folder or text input.")
+
+    # Embed and store in ChromaDB
+    embedding_fn = SentenceTransformerEmbeddingFunction(model_name="all-MiniLM-L6-v2")
+    client = chromadb.Client()
+    collection = client.get_or_create_collection(
+        name="bridgekit_ask",
+        embedding_function=embedding_fn
+    )
+    collection.add(
+        documents=chunks,
+        ids=[f"chunk_{i}" for i in range(len(chunks))]
+    )
+
+    # Retrieve most relevant chunks
+    results = collection.query(query_texts=[question], n_results=min(8, len(chunks)))
+    context = "\n\n".join(results["documents"][0])
+
+    # Generate answer with Claude
+    anthropic_client = anthropic.Anthropic(api_key=api_key)
+    message = anthropic_client.messages.create(
+        model="claude-opus-4-5",
+        max_tokens=1024,
+        system=(
+            "You are a senior data scientist answering questions based on analysis reports. "
+            "Answer only from the provided context. Be specific and cite findings where relevant. "
+            "If the context does not contain enough information to answer, say so clearly."
+        ),
+        messages=[{
+            "role": "user",
+            "content": f"Context from analysis reports:\n\n{context}\n\nQuestion: {question}"
+        }]
+    )
+
+    return message.content[0].text

bridgekit-0.1.1/README.md → bridgekit-0.2.0/bridgekit.egg-info/PKG-INFO

@@ -1,3 +1,28 @@
+Metadata-Version: 2.4
+Name: bridgekit
+Version: 0.2.0
+Summary: AI tools that make you a better data scientist, not a redundant one.
+License: MIT
+Project-URL: Homepage, https://github.com/getbridgekit/bridgekit
+Project-URL: Issues, https://github.com/getbridgekit/bridgekit/issues
+Keywords: data science,AI,analysis,evaluation,anthropic
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: anthropic>=0.20.0
+Requires-Dist: chromadb>=0.4.0
+Requires-Dist: sentence-transformers>=2.0.0
+Requires-Dist: pypdf>=3.0.0
+Requires-Dist: python-docx>=1.0.0
+Requires-Dist: python-pptx>=0.6.0
+Requires-Dist: nbformat>=5.0.0
+Dynamic: license-file
+
 # Bridgekit
 
 **AI tools that make you a better data scientist, not a redundant one.**
@@ -115,6 +140,47 @@ Paste your writeup as a string and call `evaluate()` — that's it.
 
 ---
 
+## Tool #2: Analysis Search
+
+Ask questions across a collection of your past analysis documents. Point it at a folder and get answers grounded in your actual work — no digging through files manually.
+
+Uses a vector database and semantic similarity to find relevant context across your documents — not keyword matching.
+
+Supports `.txt`, `.md`, `.pdf`, `.docx`, `.pptx`, and `.ipynb` files.
+
+**From a folder:**
+```python
+from bridgekit import ask
+
+print(ask("what drove churn in Q3?", source="reports/"))
+```
+
+**From raw text:**
+```python
+from bridgekit import ask
+
+text = """
+Q3 churn rose to 4.5%, driven by a product outage in August and a pricing
+change in July that increased SMB costs by 12%.
+"""
+
+print(ask("what caused the Q3 churn spike?", text=text))
+```
+
+**Output** *(based on sample data included in the repo)*:
+```
+Based on the Q3 2024 Churn Analysis, two primary factors drove the elevated
+churn rate of 4.5%:
+
+1. August Product Outage — A 14-hour outage affected 3,800 accounts. Impacted
+   accounts churned at 8.1% vs 3.2% for unaffected accounts.
+
+2. July Pricing Change — SMB costs increased by an average of 12%, causing SMB
+   churn to spike to 7.2% — the highest single-month figure in the dataset.
+```
+
+---
+
 ## Why not just use Claude?
 
 You could. But you'd need to know what to ask, how to frame it, and what a good answer looks like. Bridgekit has that baked in — it knows you're a data scientist presenting findings, so it asks the right questions automatically. No prompt engineering required. Just paste your work and run it.
@@ -137,7 +203,7 @@ Bridgekit only ever sees text you write yourself — your narrative, your conclu
 
 ## What's next?
 
-Bridgekit is a suite, not a one-off. The analysis reviewer is tool #1. Coming next:
+Bridgekit is a suite, not a one-off. Two tools are live — more are coming:
 
 - **Statistical approach suggester** — describe your problem in plain English, get the right test and why
 - **Stakeholder translator** — turn your technical findings into a narrative a non-technical audience will actually follow

{bridgekit-0.1.1 → bridgekit-0.2.0}/bridgekit.egg-info/SOURCES.txt

@@ -3,6 +3,7 @@ README.md
 pyproject.toml
 bridgekit/__init__.py
 bridgekit/reviewer.py
+bridgekit/search.py
 bridgekit.egg-info/PKG-INFO
 bridgekit.egg-info/SOURCES.txt
 bridgekit.egg-info/dependency_links.txt

bridgekit-0.2.0/bridgekit.egg-info/requires.txt

@@ -0,0 +1,7 @@
+anthropic>=0.20.0
+chromadb>=0.4.0
+sentence-transformers>=2.0.0
+pypdf>=3.0.0
+python-docx>=1.0.0
+python-pptx>=0.6.0
+nbformat>=5.0.0

{bridgekit-0.1.1 → bridgekit-0.2.0}/pyproject.toml

@@ -2,9 +2,12 @@
 requires = ["setuptools>=61.0"]
 build-backend = "setuptools.build_meta"
 
+[tool.setuptools.packages.find]
+include = ["bridgekit*"]
+
 [project]
 name = "bridgekit"
-version = "0.1.1"
+version = "0.2.0"
 description = "AI tools that make you a better data scientist, not a redundant one."
 readme = "README.md"
 requires-python = ">=3.9"
@@ -19,6 +22,12 @@ classifiers = [
 ]
 dependencies = [
     "anthropic>=0.20.0",
+    "chromadb>=0.4.0",
+    "sentence-transformers>=2.0.0",
+    "pypdf>=3.0.0",
+    "python-docx>=1.0.0",
+    "python-pptx>=0.6.0",
+    "nbformat>=5.0.0",
 ]
 
 [project.urls]

bridgekit-0.1.1/bridgekit/__init__.py

@@ -1,4 +0,0 @@
-from .reviewer import evaluate
-
-__version__ = "0.1.1"
-__all__ = ["evaluate"]

bridgekit-0.1.1/bridgekit.egg-info/requires.txt

@@ -1 +0,0 @@
-anthropic>=0.20.0