🫂Explore the Architecture

^{🧙‍♂️ For the ones keen to understand deeper what happens under the hood... The core of Defi Wizard is a highly modular agent swarm architecture, where each component plays a specialized role. Powered by n8n, this system is orchestrated to maximize data quality, minimize latency, and deliver sharp, multi-layered DeFi insights within seconds.}

Defi Wizard Agent Swarm

8MB

Agent Swarm Example.mp4

Watch the magic happening behind the scenes for a better visualization of its power.

---

🧱 1. The Frontliner

🔍 Purpose: Serves as the gateway for resolving any token query, enriching context for downstream agents.

🛠️ Core Functions:

• Resolves token identity via name, ticker, or contract address

• Validates network compatibility across 180+ chains

• Fetches token metadata: market cap, 24h volume, creation date, socials, and chain info

Guarantees all downstream agents receive high-integrity, pre-validated data, increasing system speed, accuracy, and coherence.

---

🧙‍♂️ 2. The Wizard (Master Orchestrator)

🧠 Purpose: Acts as the central AI coordinator — the "brain" of the swarm.

⚙️ Core Functions:

• Interprets user intent

• Delegates subtasks to relevant sub-agents

• Synthesizes and formats multi-dimensional insights into actionable responses

• Delivers HTML-rich Telegram clean outputs that are formatted with valid links

⚡ Key Features:

• Smart task allocation to reduce redundant API calls

• Balances CEX, DEX, technical, and social data into unified responses

• Built-in rate limit intelligence to avoid API lockouts

---

🧩 3. Guild of specialized Sub-Wizards-Agents

Each sub-agent focuses on a specific data domain and executes tasks in parallel

🏦 CEX Sub-Agent

• Fetches CEX data (price, volume, trends) from CoinGecko MCP

• Specializes in blue-chip tokens and widely-listed assets

🦄 DEX Sub-Agent

• Retrieves DEX token and pool data across 200+ chains and 1,600+ DEXs

• Capable of batching up to 30 tokens per API call, ensuring scalability

📊 TA Agent Master

• Conducts multi-timeframe technical analysis

• Splits into 3 sub-agents to handle multiple timeframes (e.g., 15m, 1h, 4h, 1d) simultaneously

• Uses GeckoTerminal + Syve.ai for OHLCV data

  • Up to 2,500 candles (Base/Ethereum)

  • Up to 1,000 candles or 6 months for other chains

🐦 Twitter X Agent

• Evaluates Twitter activity for influencer behavior, community engagement, and sentiment quality

• Flags red flags like bot-like engagement, giveaway spam, or fake blue checks

---

🔮 Future Sub-Agents (Planned)

• 🛡 Token Security Specialist – Automated contract auditing

• 🌍 Social Sentiment Analyst – Cross-platform narrative detection

---

⚙️ Workflow Precision

🧪 Metadata Enrichment Frontliner ensures all agents begin with a clean, structured context, avoiding overlap and boosting task accuracy.

🧠 Memory Management Maintains limited conversational memory for session continuity. (Session context engine upgrade in progress)

🔁 Robust Error Handling Retries and fallback logic for failed API calls ensures response consistency.

📉 Rate Limit Optimization Smart request pacing and batching keep all agents within their provider limits