ragcheck-cli 0.2.0__tar.gz → 0.2.2__tar.gz

{ragcheck_cli-0.2.0 → ragcheck_cli-0.2.2}/PKG-INFO

@@ -1,193 +1,195 @@
-Metadata-Version: 2.4
-Name: ragcheck-cli
-Version: 0.2.0
-Summary: Lighthouse for RAG systems — diagnose and fix your retrieval pipeline
-Project-URL: Homepage, https://github.com/pranay7863/ragcheck
-Project-URL: Documentation, https://github.com/pranay7863/ragcheck/blob/main/README.md
-Project-URL: Repository, https://github.com/pranay7863/ragcheck
-Project-URL: Issues, https://github.com/pranay7863/ragcheck/issues
-Author-email: Pranay Mane <pranaymane78@gmail.com>
-License: MIT
-License-File: LICENSE
-Keywords: ai,chunking,diagnostics,evaluation,llm,rag,retrieval
-Classifier: Development Status :: 3 - Alpha
-Classifier: Intended Audience :: Developers
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.11
-Classifier: Programming Language :: Python :: 3.12
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Classifier: Topic :: Software Development :: Quality Assurance
-Requires-Python: >=3.10
-Requires-Dist: chromadb>=0.4.0
-Requires-Dist: jinja2>=3.1.0
-Requires-Dist: litellm>=1.0.0
-Requires-Dist: nltk>=3.9.0
-Requires-Dist: pydantic>=2.5.0
-Requires-Dist: pyyaml>=6.0
-Requires-Dist: rich>=13.0.0
-Requires-Dist: sentence-transformers>=2.2.0
-Requires-Dist: transformers>=4.30.0
-Requires-Dist: typer>=0.12.0
-Provides-Extra: export
-Requires-Dist: playwright>=1.40.0; extra == 'export'
-Provides-Extra: pdf
-Requires-Dist: pypdf2>=3.0.0; extra == 'pdf'
-Provides-Extra: ragas
-Requires-Dist: ragas<0.5.0,>=0.4.0; extra == 'ragas'
-Description-Content-Type: text/markdown
-
-# ragcheck - Lighthouse for RAG Systems
-
-[![PyPI version](https://badge.fury.io/py/ragcheck.svg)](https://badge.fury.io/py/ragcheck)
-[![Python](https://img.shields.io/badge/python-3.11%2B-blue)](https://www.python.org/)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-
-> One command to diagnose your RAG pipeline and get actionable fixes.
-
-```bash
-pip install ragcheck
-ragcheck init
-ragcheck run --docs ./data --query "What is Article 370?"
-```
-
-## What is ragcheck?
-
-**ragcheck** is a lightweight, one-command diagnostic CLI that generates a beautiful, shareable HTML report analyzing why your RAG system fails and how to fix it.
-
-Think of it as **Lighthouse for RAG systems** — just like Lighthouse audits web pages, ragcheck audits your retrieval pipeline.
-
-## Features
-
-- **Auto-Generated Test Suite** - 50 synthetic questions from your documents
-- **Chunk Visualizer** - See exactly where your chunking breaks
-- **Retrieval Heatmap** - Identify dead chunks and dominant chunks
-- **Failure Classification** - Know WHY your RAG fails, not just THAT it fails
-- **Actionable Recommendations** - Specific fixes with predicted impact
-- **CI/CD Integration** - Fail builds when RAG quality regresses
-
-## Quick Start
-
-### Installation
-
-```bash
-pip install ragcheck
-```
-
-Or with [uv](https://github.com/astral-sh/uv):
-
-```bash
-uv tool install ragcheck
-```
-
-### Initialize
-
-```bash
-ragcheck init
-```
-
-Creates a `ragcheck.yaml` config file in your project.
-
-### Run Analysis
-
-```bash
-ragcheck run --docs ./data --query "Your test query"
-```
-
-Generates `ragcheck_report.html` with:
-- Scorecards (retrieval accuracy, faithfulness)
-- Chunk boundary visualization
-- Retrieval heatmap
-- Failure mode classification
-- Before/after score predictions
-
-### CI Mode
-
-```bash
-ragcheck run --docs ./data --ci --min-score 0.80
-```
-
-Returns exit code 0/1. Use in GitHub Actions to fail builds on quality regression.
-
-## Example Report
-
-![ragcheck report](https://raw.githubusercontent.com/yourusername/ragcheck/main/docs/report-screenshot.png)
-
-## Architecture
-
-```
-ragcheck CLI
-    ├── Chunk Analyzer (6 strategies + benchmark)
-    ├── Retriever Tester (auto-QA + dense retrieval)
-    ├── Failure Classifier (4 failure modes)
-    ├── Recommendation Engine (decision tree)
-    └── Report Engine (Jinja2 + CSS/HTML HTML)
-```
-
-## Tech Stack
-
-| Component | Tool |
-|-----------|------|
-| CLI | Typer + Rich |
-| Config | Pydantic |
-| Embeddings | sentence-transformers |
-| Vector DB | ChromaDB |
-| LLM Interface | LiteLLM |
-| Reports | Jinja2 + CSS/HTML |
-
-## Configuration
-
-`ragcheck.yaml`:
-
-```yaml
-project_name: ragcheck
-docs_path: ./data
-chunking:
-  strategy: recursive
-  chunk_size: 512
-  chunk_overlap: 128
-llm:
-  provider: openai
-  model: gpt-3.5-turbo
-retrieval:
-  top_k: 5
-  similarity_threshold: 0.7
-report:
-  format: html
-  include_heatmap: true
-```
-
-## Development
-
-```bash
-git clone https://github.com/pranay7863/ragcheck.git
-cd ragcheck
-uv sync
-uv run pytest
-uv run ruff check .
-uv run mypy ragcheck/
-```
-
-## Contributing
-
-See [CONTRIBUTING.md](CONTRIBUTING.md)
-
-## License
-
-MIT — see [LICENSE](LICENSE)
-
-## Roadmap
-
-- [x] v0.2.0 — Offline reports, NLI faithfulness, scaled auto-QA, chunk viz
-- [ ] v0.3.0 — More vector DBs (Pinecone, Weaviate)
-- [ ] v0.3.0 — SaaS API for teams
-- [ ] v0.4.0 — Enterprise features (SSO, audit logs)
-
-## Support
-
-- [GitHub Issues](https://github.com/pranay7863/ragcheck/issues)
-- Twitter: [@ypranay53](https://twitter.com/pranay53)
-
----
-
-**Built with discipline.** Read the [blueprint](docs/ARCHITECTURE.md) that started it all.
+Metadata-Version: 2.4
+Name: ragcheck-cli
+Version: 0.2.2
+Summary: Lighthouse for RAG systems - diagnose and fix your retrieval pipeline
+Author-email: Pranay Mane <pranaymane78@gmail.com>
+License: MIT
+Project-URL: Homepage, https://github.com/pranay7863/ragcheck
+Project-URL: Documentation, https://github.com/pranay7863/ragcheck/blob/main/README.md
+Project-URL: Repository, https://github.com/pranay7863/ragcheck
+Project-URL: Issues, https://github.com/pranay7863/ragcheck/issues
+Keywords: rag,llm,evaluation,retrieval,ai,chunking,diagnostics
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Software Development :: Quality Assurance
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: typer>=0.12.0
+Requires-Dist: rich>=13.0.0
+Requires-Dist: jinja2>=3.1.0
+Requires-Dist: nltk>=3.9.0
+Requires-Dist: pydantic>=2.5.0
+Requires-Dist: litellm>=1.0.0
+Requires-Dist: sentence-transformers>=2.2.0
+Requires-Dist: chromadb>=0.4.0
+Requires-Dist: PyYAML>=6.0
+Requires-Dist: transformers>=4.30.0
+Requires-Dist: pdfplumber>=0.10.0
+Provides-Extra: pdf
+Requires-Dist: PyPDF2>=3.0.0; extra == "pdf"
+Provides-Extra: export
+Requires-Dist: playwright>=1.40.0; extra == "export"
+Provides-Extra: ragas
+Requires-Dist: ragas<0.5.0,>=0.4.0; extra == "ragas"
+Dynamic: license-file
+
+# ragcheck - Lighthouse for RAG Systems
+
+[![PyPI version](https://badge.fury.io/py/ragcheck-cli.svg)](https://badge.fury.io/py/ragcheck-cli)
+[![Python](https://img.shields.io/badge/python-3.11%2B-blue)](https://www.python.org/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+
+> One command to diagnose your RAG pipeline and get actionable fixes.
+
+```bash
+pip install ragcheck-cli
+ragcheck init
+ragcheck run --docs ./data --query "What is Article 370?"
+```
+
+## What is ragcheck?
+
+**ragcheck** is a lightweight, one-command diagnostic CLI that generates a beautiful, shareable HTML report analyzing why your RAG system fails and how to fix it.
+
+Think of it as **Lighthouse for RAG systems** — just like Lighthouse audits web pages, ragcheck audits your retrieval pipeline.
+
+## Features
+
+- **Auto-Generated Test Suite** - 50 synthetic questions from your documents
+- **Chunk Visualizer** - See exactly where your chunking breaks
+- **Retrieval Heatmap** - Identify dead chunks and dominant chunks
+- **Failure Classification** - Know WHY your RAG fails, not just THAT it fails
+- **Actionable Recommendations** - Specific fixes with predicted impact
+- **CI/CD Integration** - Fail builds when RAG quality regresses
+
+## Quick Start
+
+### Installation
+
+```bash
+pip install ragcheck-cli
+```
+
+Or with [uv](https://github.com/astral-sh/uv):
+
+```bash
+uv tool install ragcheck-cli
+```
+
+### Initialize
+
+```bash
+ragcheck init
+```
+
+Creates a `ragcheck.yaml` config file in your project.
+
+### Run Analysis
+
+```bash
+ragcheck run --docs ./data --query "Your test query"
+```
+
+Generates `ragcheck_report.html` with:
+- Scorecards (retrieval accuracy, faithfulness)
+- Chunk boundary visualization
+- Retrieval heatmap
+- Failure mode classification
+- Before/after score predictions
+
+### CI Mode
+
+```bash
+ragcheck run --docs ./data --ci --min-score 0.80
+```
+
+Returns exit code 0/1. Use in GitHub Actions to fail builds on quality regression.
+
+## Example Report
+
+![ragcheck report](https://raw.githubusercontent.com/pranay7863/ragcheck/main/docs/report-screenshot.png)
+
+## Architecture
+
+```
+ragcheck CLI
+    ├── Chunk Analyzer (6 strategies + benchmark)
+    ├── Retriever Tester (auto-QA + dense retrieval)
+    ├── Failure Classifier (4 failure modes)
+    ├── Recommendation Engine (decision tree)
+    └── Report Engine (Jinja2 + CSS/HTML)
+```
+
+## Tech Stack
+
+| Component | Tool |
+|-----------|------|
+| CLI | Typer + Rich |
+| Config | Pydantic |
+| Embeddings | sentence-transformers |
+| Vector DB | ChromaDB |
+| LLM Interface | LiteLLM |
+| Reports | Jinja2 + CSS/HTML |
+
+## Configuration
+
+`ragcheck.yaml`:
+
+```yaml
+project_name: ragcheck
+docs_path: ./data
+chunking:
+  strategy: recursive
+  chunk_size: 512
+  chunk_overlap: 128
+llm:
+  provider: openai
+  model: gpt-3.5-turbo
+retrieval:
+  top_k: 5
+  similarity_threshold: 0.7
+report:
+  format: html
+  include_heatmap: true
+```
+
+## Development
+
+```bash
+git clone https://github.com/pranay7863/ragcheck.git
+cd ragcheck
+uv sync
+uv run pytest
+uv run ruff check .
+uv run mypy ragcheck/
+```
+
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md)
+
+## License
+
+MIT — see [LICENSE](LICENSE)
+
+## Roadmap
+
+- [x] v0.2.0 — Offline reports, NLI faithfulness, scaled auto-QA, chunk viz
+- [ ] v0.3.0 — More vector DBs (Pinecone, Weaviate)
+- [ ] v0.3.0 — SaaS API for teams
+- [ ] v0.4.0 — Enterprise features (SSO, audit logs)
+
+## Support
+
+- [GitHub](https://github.com/pranay7863/ragcheck)
+- Twitter: [@mane_pranay](https://twitter.com/mane_pranay)
+
+---
+
+**Built with discipline.** Read the [blueprint](docs/ARCHITECTURE.md) that started it all.

{ragcheck_cli-0.2.0 → ragcheck_cli-0.2.2}/README.md

@@ -1,13 +1,13 @@
 # ragcheck - Lighthouse for RAG Systems
 
-[![PyPI version](https://badge.fury.io/py/ragcheck.svg)](https://badge.fury.io/py/ragcheck)
+[![PyPI version](https://badge.fury.io/py/ragcheck-cli.svg)](https://badge.fury.io/py/ragcheck-cli)
 [![Python](https://img.shields.io/badge/python-3.11%2B-blue)](https://www.python.org/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
 > One command to diagnose your RAG pipeline and get actionable fixes.
 
 ```bash
-pip install ragcheck
+pip install ragcheck-cli
 ragcheck init
 ragcheck run --docs ./data --query "What is Article 370?"
 ```
@@ -32,13 +32,13 @@ Think of it as **Lighthouse for RAG systems** — just like Lighthouse audits we
 ### Installation
 
 ```bash
-pip install ragcheck
+pip install ragcheck-cli
 ```
 
 Or with [uv](https://github.com/astral-sh/uv):
 
 ```bash
-uv tool install ragcheck
+uv tool install ragcheck-cli
 ```
 
 ### Initialize
@@ -72,7 +72,7 @@ Returns exit code 0/1. Use in GitHub Actions to fail builds on quality regressio
 
 ## Example Report
 
-![ragcheck report](https://raw.githubusercontent.com/yourusername/ragcheck/main/docs/report-screenshot.png)
+![ragcheck report](https://raw.githubusercontent.com/pranay7863/ragcheck/main/docs/report-screenshot.png)
 
 ## Architecture
 
@@ -82,7 +82,7 @@ ragcheck CLI
     ├── Retriever Tester (auto-QA + dense retrieval)
     ├── Failure Classifier (4 failure modes)
     ├── Recommendation Engine (decision tree)
-    └── Report Engine (Jinja2 + CSS/HTML HTML)
+    └── Report Engine (Jinja2 + CSS/HTML)
 ```
 
 ## Tech Stack
@@ -146,8 +146,8 @@ MIT — see [LICENSE](LICENSE)
 
 ## Support
 
-- [GitHub Issues](https://github.com/pranay7863/ragcheck/issues)
-- Twitter: [@ypranay53](https://twitter.com/pranay53)
+- [GitHub](https://github.com/pranay7863/ragcheck)
+- Twitter: [@mane_pranay](https://twitter.com/mane_pranay)
 
 ---
 

{ragcheck_cli-0.2.0 → ragcheck_cli-0.2.2}/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "ragcheck-cli"
-version = "0.2.0"
-description = "Lighthouse for RAG systems — diagnose and fix your retrieval pipeline"
+version = "0.2.2"
+description = "Lighthouse for RAG systems - diagnose and fix your retrieval pipeline"
 readme = "README.md"
 license = {text = "MIT"}
 requires-python = ">=3.10"
@@ -31,6 +31,7 @@ dependencies = [
     "chromadb>=0.4.0",
     "PyYAML>=6.0",
     "transformers>=4.30.0",
+	"pdfplumber>=0.10.0",
 ]
 
 [project.optional-dependencies]
@@ -48,11 +49,12 @@ Repository = "https://github.com/pranay7863/ragcheck"
 Issues = "https://github.com/pranay7863/ragcheck/issues"
 
 [build-system]
-requires = ["hatchling"]
-build-backend = "hatchling.build"
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
 
-[tool.hatch.build.targets.wheel]
-packages = ["src/ragcheck"]
+[tool.setuptools.packages.find]
+where = ["."]
+include = ["ragcheck*"]
 
 [tool.ruff]
 line-length = 100

{ragcheck_cli-0.2.0 → ragcheck_cli-0.2.2}/ragcheck/__init__.py

@@ -1,3 +1,3 @@
 """ragcheck — Lighthouse for RAG systems."""
 
-__version__ = "0.2.0"
+__version__ = "0.2.2"