Veritasgraph

VeritasGraph — The Governed, On-Prem GraphRAG & Agent Framework

Stop chunking blindly. Combine Tree-Search structure with Knowledge-Graph reasoning — and wire it into governed AI agents. Runs 100% locally or in the cloud.

🎯 Traditional RAG guesses based on similarity. VeritasGraph reasons based on structure.

Don't just find the document — understand the connection, then act on it with governed agents.

🌐 Live site · ⭐ Star · 🍴 Fork · 💬 Discuss · 🐛 Report a bug

python -m veritasgraph_mcp     # from repo root (needs local Ollama for ingest/query)

[▶️️ Watch the Full Talk & Live Demo on YouTube (27 mins)](https://www.youtube.com/watch?v=MDo1Mk8R5Wo)

Why Vector RAG Breaks on Deep Research

• Attribution is a data model, not a prompt: Storing provenance at write time prevents hallucinated citations at read time.
• Deterministic retrieval: Graph traversal is fully reproducible, separating factual evidence traces from stochastic LLM generation.
• Explicit abstention: The agent identifies missing graph edges and returns "insufficient evidence" rather than guessing.

-

📚 Featured Guide — Build Governed AI Agents On-Prem

A complete walkthrough of designing, wiring, and shipping governed AI agents entirely on your own infrastructure.

**📄 Read the guide: *Build Governed AI Agents On-Prem* (PDF)**

▶️ [Watch the walkthrough on YouTube](https://youtu.be/sA7ReEgdJfg)

🚀 Quick Start (2 lines, no GPU)

pip install veritasgraph

veritasgraph demo --mode=lite

That's it — an interactive demo using cloud APIs (OpenAI/Anthropic), no local models required.

export OPENAI_API_KEY="sk-..."        # Lite: cloud APIs, zero setup

veritasgraph demo --mode=lite

veritasgraph demo --mode=local --model=llama3.2 # 100% offline with Ollama veritasgraph start --mode=full # full GraphRAG pipeline

Useful links: ⚡ Live docs · 🎮 Live demo · 📖 Article · 📄 Research paper

🏥 Example Implementation — MedTrust Agent

[MedTrust Agent](MedTrust%20Agent/README.md) is a complete, reproducible example implementation of the VeritasGraph governed-agent architecture, applied to medical and pharmaceutical workflows — a zero-trust, observable LLM agent framework (NEECON 2026 paper).

• Zero-trust control plane: least-privilege IAM with step-up authorization — no tool runs before authorization succeeds.
• Governed answers: mandatory [doc#chunk] citations; uncited answers are withheld or routed to human review.
• Tamper-evident audit: keyed, hash-chained audit log with published anchors.
• Honest evaluation: security measured directly from the running code (100% unauthorized actions blocked, Wilson 95% CIs), answer quality clearly labelled *illustrative*; 69 tests, a versioned 40-scenario suite, and a one-command reproduce script.

python -m medtrust.demo    # end-to-end governed demo, 100% offline

📄 Read the full write-up: MedTrust Agent README

🧩 Built with VeritasGraph

A growing ecosystem of projects is building on VeritasGraph for enterprise GraphRAG, agent orchestration, fact-checking, and governed reasoning.

Community projects and integrations

• MeAkash77 / VeritasGraph-Enterprise-GraphRAG-Verifiable-AI-Platform — enterprise GraphRAG adapter and verifiable AI platform patterns.
• PTP063 / veritas-graph — derivative project and live deployment using VeritasGraph concepts.
• RahulSagar2005 / multi-hop-rag — multi-hop retrieval and reasoning workflows.
• lukehungngo / agentwall — agent security and governance-oriented reasoning flows.
• gajanan229 / Fact-checker-Agent — evidence-focused fact-checking and explanation patterns.
• Minh-Tam-Solution / EndiorBot — enterprise chatbot and AI agent integration built around graph-backed reasoning.
• NistorDorinGabriel / Proiect_Licenta — academic thesis project extending VeritasGraph-inspired graph reasoning techniques.

If you are building with VeritasGraph and want to be featured here, open a PR or reach out with a short repo description and usage summary.

� Live Example — Floorplan-KG Takeoff Demo

A hosted, zero-setup example of VeritasGraph applied to construction drawings. Upload a floorplan PDF (or pick a bundled sample) and the engine extracts rooms, dimensions, openings, fixtures, and schedules — then builds a knowledge graph and quantity takeoff you can query, all with verifiable provenance back to the source sheet.

[🏗️ Try the Floorplan-KG demo](https://spaceai-llmpc.ngrok.app/floorplan) — *free, no login; runs the live PyMuPDF/pdfplumber extraction pipeline.*

• What it shows: PDF → structured rooms/dimensions/schedules → knowledge graph → CSV/XLSX takeoff export, with a VeritasGraph assist panel for grounded Q&A over the extracted graph.
• Why it matters: the same governed-attribution model that powers document RAG applies to geometry — every measured quantity traces back to a specific sheet and element rather than a guessed value.
• Example output: a single-sheet residential plan returns ~17 rooms and ~2,400 sq ft in under 200 ms.

Studio is a local Agent Build Workspace (FastAPI + single-page UI) that lets you build a knowledge graph from your own documents and wire it into agents alongside tools, memory, data logging, guardrails, and headroom-style context budgeting — then chat with those agents live and watch every stage of the orchestration pipeline. Everything runs 100% locally against Ollama.

[🎮 Try the Studio Live](https://bibinprathap.github.io/VeritasGraph/studio/) — *stable URL that always redirects to the current running studio tunnel.*

Run it:

pip install -r requirements.txt

ollama serve & ollama pull qwen3:latest # any local chat model STUDIO_DATA_DIR="$PWD/studio_api/data" \ uvicorn studio_api.main:app --host 127.0.0.1 --port 8200 --log-level warning

Studio UI → http://localhost:8200/studio · API docs → /docs

One-command end-to-end demo (builds a graph + drives a fully-wired agent through graph reasoning, memory recall, PII redaction, and a guardrail block):

python3 demos/agent-studio/sample_pipeline.py --model qwen3:latest

What's inside — full Studio feature set

• 🧩 Knowledge Graph builder & explorer — ingest text, extract entities/relationships locally, inspect nodes/edges with grounded evidence.
• 🔎 Graph Q&A with citations — multi-hop answers backed by [doc#chunk] source attribution.
• 🤖 Agent workspace — create/edit agents with model selection, prompt/persona settings, and per-agent capability toggles.
• 🔀 Governed orchestration pipeline — per-turn flow of Guardrails → Memory → Knowledge Graph → Headroom budget → Tools → Data log, with full trace visibility.
• 🧰 Editable tools catalog — add, edit, enable/disable, test, and delete tools directly in Studio.
• 🌐 External real tool support — call real HTTP endpoints with configurable method, auth header, and custom headers.
• 🔌 MCP bridge integrations — local MCP proxy connectors (e.g. Chrome DevTools MCP, Unity MCP) with health-aware probing.
• 🛡️ Guardrails — PII redaction and policy-block controls with visible guardrail-block metrics.
• 🧠 Memory + Data logs — per-agent short-term memory and interaction-log persistence.
• 📈 Evaluation & fine-tune simulation — run eval suites, track pass-rate trends, and queue/monitor fine-tune jobs.
• 💬 Playground — run governed agent conversations live and inspect the pipeline trace.
• 📊 KPI dashboard — active agents, connected tools, eval pass rate, and guardrail-block counters.

See [`studio_api