# juanjuandog/FinSight-AI

AI equity research agent with resilient workflows, evidence-grounded RAG, versioned reports, and automated quality evaluation.

Repository: https://github.com/juanjuandog/FinSight-AI
Canonical: https://ross.abutalabs.com/products/finsight-ai
Language: Java
License: MIT
License Family: permissive
Topics: ai-agent, financial-research, llm-evaluation, pgvector, postgresql, rabbitmq, rag, redis, spring-boot, workflow-orchestration
Last push: 2026-09-02T13:54:32+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 35, longevity 8
- inputs: {"age_days": 114, "days_push": 0, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1034, forks 62 (observed 2026-09-03T02:15:13.246174+00:00)

## What it is
FinSight AI is an open-source equity research workspace for A-share companies that turns market data, filings, and financial metrics into structured, evidence-grounded AI research reports. Built on Java 17 and Spring Boot with PostgreSQL/pgvector, RabbitMQ, and Redis, it emphasizes recoverable long-running agent workflows, snapshot-bound versioned reports, and automated LLM output evaluation.

## Use cases
- research A-share companies with AI-generated analysis
- build reliable long-running LLM agent workflows
- generate evidence-grounded financial research reports
- run RAG over company filings and announcements
- evaluate LLM output quality automatically
- track company events and risk signals on a timeline
- self-host an AI equity research assistant

## When to choose
- you need reproducible, evidence-backed AI research reports rather than raw chatbot answers
- you want a production-grade reference for resilient agent workflow orchestration in Java/Spring Boot
- you research Chinese A-share equities and want filings, quotes, and metrics in one workspace
- you need versioned reports bound to data snapshots with an inspectable evidence path

## When to avoid
- you need automated trading or investment execution - it is explicitly a research aid, not advice
- you need coverage of non-A-share markets like US or European equities
- you want a lightweight Python-based LLM stack instead of a JVM/Spring Boot backend
- you need real-time streaming market data

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, rag, llm-inference, workflow-automation, search-engine, data-visualization, llm-training, monitoring
- domain: artificial-intelligence, fintech, large-language-models
- platform: jvm, self-hosted, windows
- tags: equity-research, spring-boot, pgvector, rabbitmq, redis, a-share, financial-analysis, workflow-orchestration, evidence-grounded, report-generation, ai-agents, retrieval-augmented-generation, data-engineering, docker, web-server, linux, macos

## Member repositories
- juanjuandog/FinSight-AI (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:13.246174+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:07:27.273173+00:00, confidence not recorded.
  - readme: https://github.com/juanjuandog/FinSight-AI (fetched 2026-09-03T02:15:13.246174+00:00, sha 314e0388c623)
- Data as of 2026-08-30T08:39:29.467469+00:00.