HumanValue

Talent Value Intelligence Platform
Conversational AI · Agent Tools · Automated Performance Evaluation · Multi-perspective Assessment

Read this in: English · 中文 · 日本語

---

HumanValue is an AI-driven platform that quantifies and grows talent value. It combines a conversational AI agent, a full agent toolchain, and automated multi-perspective evaluation — designed for managers and HR to understand, develop, and maximize the value of every person. The UI is English-first with switchable 中文 / 日本語.

---

Highlights

• Conversational AI & Agent Console — an AI assistant that is also a control console: run analyses, create announcements, manage tickets, run pipelines, take backups, and more, directly in chat.
• Talent Value Engine — 10 theory-driven analyses: 9-box classification, key-person / single-point-of-failure risk, Pareto concentration, team efficiency, incentive strategy, compensation competitiveness, succession pipeline, burnout warning, skill-fit & reallocation, and quarterly review.
• Multi-System-Type — the same engine adapts to different institutional contexts: enterprise (elimination-based), academia (cultivation-based), public sector (promotion-based), training (certification-based), and gig platforms (flexible).
• Enterprise Governance — MFA, login brute-force protection, announcements, tickets, data assets, data pipelines, disaster recovery backups, and AI security red-team.
• Universal Agent Capabilities — ReAct loop, planner, reflector, multi-agent supervisor, MCP client/server, A2A protocol, browser automation, skills, memory & RAG, streaming SSE, context compression, and more.

Product Preview

Talent Dashboard	Talent Value Engine	AI Assistant (Chat Console)

Command Palette	Create Announcement (via Chat)	Dark Mode

See the full walkthrough in the Demo Showcase.

Architecture

• Frontend: Vue 3 · Pinia · Element Plus · ECharts · KaTeX · Mermaid
• Backend: Python 3.11+ · FastAPI · SQLAlchemy · Alembic
• Agent: LangGraph · ReAct · LangChain tools · MCP · A2A
• LLM Providers: OpenAI / Anthropic / Gemini / DeepSeek / Qwen / Ollama (encrypted credentials, load-balanced)
• Storage: SQLite (dev) / PostgreSQL (prod) · ChromaDB · Redis · MinIO
• Observability: Prometheus · Langfuse · Grafana · Loki

Quick Start

Docker Compose (recommended)

git clone https://gitcode.com/badhope/HumanValue.git
cd HumanValue
cp backend/.env.example backend/.env   # set JWT_SECRET_KEY and a model API key
docker compose up -d --build

Local Development

# backend
cd backend && pip install -r requirements.txt
uvicorn main:app --reload --port 8000

# frontend
cd frontend && npm install && npm run dev   # http://localhost:5173

No API Key? Mock Mode

cd backend && uvicorn main:app --reload --port 8000
python -m eval.evaluate --mock

When no LLM API key is configured, the system falls back to a deterministic Mock Provider so the full pipeline runs end-to-end.

Demo

See the Demo Showcase for the complete recorded walkthrough, screenshots, and the promotional GIF.

Documentation

Document	Description
Demo Showcase	Recorded video + screenshots
Universal Agent Capabilities	Full universal agent feature checklist
Talent Value Engine	Capabilities & APIs
Chat Console Guide	How to operate the app via conversation
Shortcuts & Tips	Global shortcuts & usage tips
Error Codes	Unified error code handbook
Agent Pitfalls	Common pitfalls & regression gates
Launch Checklist	Pre-release verification
Changelog	Version history

Repositories

Platform	URL
GitCode (primary)	https://gitcode.com/badhope/HumanValue
GitHub (mirror)	https://github.com/weed33834/HumanValue
Gitee (mirror)	https://gitee.com/badhope/HumanValue

License

Apache License 2.0. See LICENSE.