rishi-is-cool/leaselens-india

How rishi-is-cool/leaselens-india works

LeaseLens accepts lease files and currently returns extracted, segmented clauses; the sampled frontend identifies itself as a Phase 1 segmentation check. The backend also contains a document-explanation pipeline for risk labels, jurisdiction/topic-scoped statute retrieval, grounded explanations, and cross-clause connections. The README describes Postgres-backed document and statute storage. The graph follows calls visible in sampled source and documented capabilities; unsampled frontend API wiring and analysis internals are represented only where their paths and responsibilities are supported.

Main entry points: documents.py Document endpoints · pipeline.py Explanation orchestration · App.tsx Upload and results UI · api.py Risk classification

Components 22

LeaseLens frontend

Upload and results UI frontend/src/App.tsx
Frontend API client frontend/src/api.ts

API and ingestion

API application backend/app/main.py
Document endpoints backend/app/routers/documents.py
Text extraction backend/app/ingestion/extract.py
Clause segmentation backend/app/ingestion/segment.py
Upload retention backend/app/retention.py

Clause analysis

Explanation orchestration backend/app/explain/pipeline.py
Risk classification backend/app/classifier/api.py
Trained fallback backend/app/classifier/trained.py
LLM classifier backend/app/classifier/llm.py
Topic tagging backend/app/explain/topics.py
Statute retrieval backend/app/statute_kb/retrieval.py
Explanation generation backend/app/explain/generate.py
Citation grounding backend/app/explain/grounding.py
Cross-clause analysis backend/app/explain/cross_clause.py
Shared LLM client backend/app/llm_client.py

Persistence and providers

Database access backend/app/db.py
Document and clause models backend/app/models.py
Postgres

Other components

Lease uploader
LLM provider

Connections 23

Lease uploader → Upload and results UI: uploads lease
Upload and results UI → Frontend API client: sends file
Frontend API client → Document endpoints: calls upload API
API application → Document endpoints: registers routes
Document endpoints → Upload retention: purges expired
Document endpoints → Text extraction: extracts text
Document endpoints → Clause segmentation: segments paragraphs
Document endpoints → Database access: opens session
Document endpoints → Document and clause models: creates records
Database access → Postgres: reads and writes
Upload and results UI → Frontend API client: checks health
Explanation orchestration → Risk classification: classifies clauses
Explanation orchestration → Topic tagging: tags topics
Explanation orchestration → Statute retrieval: retrieves statutes
Explanation orchestration → Explanation generation: generates explanations
Explanation orchestration → Citation grounding: checks grounding
Explanation orchestration → Cross-clause analysis: checks clause pairs
Risk classification → Trained fallback: uses fallback
Risk classification → LLM classifier: uses LLM
Explanation generation → Shared LLM client: requests completion
Cross-clause analysis → Shared LLM client: requests completion
Shared LLM client → LLM provider: sends chat request
LLM classifier → LLM provider: sends chat request

More Python diagrams: imlrz/deepresearch-bench-ii · jeankossaifi/fista · jinef-john/google-maps-scraper · jingliinpurdue/fast-and-robust-uav-to-uav-detection-and-tracking · lewjiayi/crowd-analysis

Drawn by GitDiagram on September 29, 2026 from the repository's file tree, README and source files. Read as Markdown ·