> For the complete documentation index, see [llms.txt](https://docs.hermiv.tech/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.hermiv.tech/getting-started/overview.md).

# Architecture Overview

Hermiv is architected around a strict **Decoupled Dual-Pipeline** pattern that completely isolates ultra-low-latency bot execution (**Fast-Path**) from heavy telemetry enrichment and human notification (**Slow-Path**).

```
                             [ RAW ON-CHAIN INGEST ]
                   Phantom SSE (Solana) • WS Trenches (RH / BSC)
                                       │
                                       ▼
                       ┌───────────────────────────────┐
                       │    PRE-FLIGHT VALIDATION      │
                       │ • On-Chain Bytecode Vaccine   │
                       │ • Trader Blacklist (O(1))     │
                       │ • Min USD Threshold Filter    │
                       └───────────────┬───────────────┘
                                       │
             ┌─────────────────────────┴─────────────────────────┐
             ▼ (< 0.1 ms Latency Budget)                         ▼ (Detached Background Async)
   ⚡ FAST-PATH ENGINE                                 🐢 SLOW-PATH WORKER
   • Canonical Transformer (`alpha.v1`)                • Token-Deduplicated Cache (30s window)
   • Unified WebSocket Gateway (`/v1/ws`)              • DEXScreener Pair & Liquidity Fetch
   • Multiplexed SSE Streams (`/v1/stream`)            • GMGN KOL & Smart Holder Intelligence
   • In-Memory Event Bus (`SignalHub`)                 • Rate-Limited Discord Queue Serializer
   • Sub-millisecond Bot Dispatches                    • Multi-Channel Alert Formatting
```

***

## ⚡ 1. The Fast-Path (< 0.1 ms)

The Fast-Path is dedicated entirely to **algorithmic trading bots and execution daemons**:

* **Zero External HTTP Network Delays:** The engine never awaits external third-party HTTP endpoints (DEXScreener, GMGN, CoinGecko) during the fast-path loop. All processing runs purely in memory.
* **Deterministic Execution:** Once a swap is validated against our in-memory honeypot bytecode cache and trader blacklist, it is instantly converted to the standard `alpha.v1` schema.
* **Non-Blocking Fanout:** Sockets receive the raw buffer write immediately. If a consumer socket drops or encounters network backpressure, it is purged without blocking sibling connections.

***

## 🐢 2. The Slow-Path (Enrichment & Forensics)

The Slow-Path handles heavy off-chain analytical intelligence for human traders and monitoring dashboards:

* **Token-Deduplicated Worker:** When an aggressive accumulation wave occurs (e.g. 5 smart wallets dogpiling a newly bonded coin within 30 seconds), the engine issues only **1 query** to DEXScreener and GMGN, reusing the metadata across all alerts to prevent event-loop starvation and HTTP 429 rate limits.
* **Background Isolation:** Any failure in the slow-path (e.g. Discord API outage or network timeout) has zero impact on live WebSocket and SSE consumers.

***

## 🔒 3. Enterprise Hardening & Isolation

* **Edge Armor:** Global DDoS protection and TLS 1.3 encryption terminating at the Cloudflare edge.
* **Memory Bounds:** All in-memory caches enforce strict LRU eviction ceilings (`MAX_CACHE_SIZE = 1,000`) to guarantee stability over weeks and months of continuous operation.
* **No Leaky Abstractions:** Upstream vendor endpoints and private data sources are completely masked behind the **Hermiv Engine** identity.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.hermiv.tech/getting-started/overview.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.