Multi-agent AI · Real-time streaming

Uncover the causes behind
history's turning points

Storming Bastille maps causal chains between historical events — not just facts, but the relationships between them. Five AI agents coordinate to produce a streamed narrative, an interactive causal graph, a zoomable timeline, and verified source citations.

► Try Live Demo ★ View on GitHub
5
Specialist AI agents
<2s
Time to first graph nodes
4
Orchestration phases
5
Causal edge types
SSE
Real-time streaming
History isn't linear. Your tools should reflect that.

You ask "What led to the French Revolution?" and within seconds Storming Bastille starts building a structured causal map: a streamed narrative backed by live web research, an interactive graph of connected events with confidence-weighted edges, a zoomable D3 timeline, and source citations ranked by reliability. Five AI agents coordinate behind the scenes, each doing one job well. The result is a session you can explore, edit, name, and revisit — not a paragraph you copy and paste.

Everything you need to understand causation

Built around a single idea: causal relationships between events matter more than chronological facts.

◈

Interactive Causal Graph

React Flow canvas with five relationship types — direct cause, contributing factor, enabling condition, consequence, feedback loop — each with a confidence score. Drag, edit, and export to PNG.

↻

Real-time Streaming

Narrative text streams token by token via SSE. Graph nodes appear within 2 seconds. Causal edges arrive as a second pass. You watch the analysis build itself — never stare at a spinner.

✓

Verified Source Citations

A dedicated SourceVerifier agent cross-checks claims and assigns reliability tiers: Academic, Encyclopedia, News, Web, Blog. Citations appear in a ranked panel alongside the graph.

◷

D3 Zoomable Timeline

Events plotted chronologically on a D3 SVG timeline. Zoomable, pannable, and linked to the graph — click a timeline event to highlight its corresponding node.

⊞

Session Memory

Name and save any analysis. Sessions persist server-side with SQLite and automatic GCS backup. Pick up where you left off, or run follow-up queries that build on prior context.

⬡

Multi-agent Pipeline

Five specialists: HistorianAgent (Haiku), CausalAnalystAgent (Sonnet), GraphBuilderAgent, SourceVerifierAgent, and FollowUpAgent. Each optimised for its precise task.

4-phase orchestration

Designed to minimise time-to-first-pixel. Something meaningful is on screen within 2 seconds of submitting a query.

1
Parallel · ~1–2s first token

Research & Narrative

DuckDuckGo and Bing are queried simultaneously. Results are injected as plain text into the HistorianAgent prompt — not via tool-use, to avoid latency overhead. HistorianAgent (Claude Haiku 4.5) streams the narrative and embeds 8–10 structured events as JSON.

2
Immediate · ~2s

Graph Nodes

GraphBuilderAgent converts the historian's structured events into React Flow nodes and Dagre-positioned timeline data. An SSE event fires — nodes appear on screen before any causal analysis begins. No AI call needed; this is pure deterministic transformation.

3
Async · 5–15s

Causal Analysis

CausalAnalystAgent (Claude Sonnet) identifies relationships between events: type, confidence score (0.3–1.0), and explanation. A second SSE event fires with the edge set. The graph completes visually as edges appear between the existing nodes.

4
Fire-and-forget

Source Verification & Persistence

SourceVerifierAgent cross-checks claims and assigns reliability tiers. Results are saved to SQLite with a GCS backup. This runs in the background — the user has everything they need before this phase completes.

Built on modern, production-grade tools

A carefully chosen stack with no unnecessary dependencies.

Backend

Python 3.12 FastAPI sse-starlette aiosqlite Anthropic SDK httpx BeautifulSoup4 PyJWT & bcrypt

Frontend

Next.js 15.5 React 19 TypeScript TailwindCSS 4 Zustand 5 React Flow v12 D3.js 7 Framer Motion

AI Models

Claude Haiku 4.5 Claude Sonnet

Infrastructure

GCP Cloud Run Artifact Registry Secret Manager Cloud Storage Terraform GitHub Actions Docker

Security Scanning

Bandit SAST pip-audit CodeQL Trivy
Running locally in under 5 minutes

Prerequisites: Python 3.12+, uv, and Bun.

1

Clone the repository

Pull the source code and enter the project directory.

2

Configure environment

Copy the backend env file and set your Anthropic API key. That's the only required configuration to start.

3

Start the application

The start.sh script boots both backend and frontend. Default login: admin / changeme.

# 1. Clone
$ git clone https://github.com/rupakc/storming-bastille.git
$ cd storming-bastille

# 2. Configure
$ cp backend/.env.example backend/.env
$ echo "ANTHROPIC_API_KEY=sk-ant-..." >> backend/.env

# 3. Run
$ ./start.sh

# Open → http://localhost:3000  (login: admin / changeme)

Full setup instructions including GCP deployment in the Setup and Installation wiki page.

Ready to map history's causes?

Clone the repo, set your API key, and start exploring any historical question in seconds.

► Try Live Demo ★ Star on GitHub
Where to go next

Everything you need to understand, extend, and deploy the project.