# gauranshahuja/jarvis architecture

> Architecture diagram of the GitHub repository gauranshahuja/jarvis, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-18). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/gauranshahuja/jarvis
- Repository: https://github.com/gauranshahuja/jarvis

## Overview

J.A.R.V.I.S. is an offline-first Windows voice assistant. The main workflow captures speech, recognizes English or Hindi, dispatches commands to skills, and speaks results; unmatched questions can use offline web answers or the optional Claude brain. Separate capabilities include biometric enrollment, persistent reminders, proactive speech, and the HUD. Gesture-control experiences are documented, but their implementation is not present in the supplied tree. Edges rely on shown excerpts, SOURCE INDEX references, or README statements; unsampled wiring is omitted where it cannot be established.

Main entry points: [`jarvis.py`](https://github.com/gauranshahuja/jarvis/blob/main/jarvis.py) (Assistant App), [`skills.py`](https://github.com/gauranshahuja/jarvis/blob/main/skills.py) (Voice Skills), [`paths.py`](https://github.com/gauranshahuja/jarvis/blob/main/paths.py) (Path Resolution), [`brain.py`](https://github.com/gauranshahuja/jarvis/blob/main/brain.py) (Command Brain)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_interaction["Voice Interaction"]
  node_app["Assistant App<br/>[jarvis.py]"]
  node_speech["Speech I/O<br/>[core.py]"]
  node_routing["Command Brain<br/>[brain.py]"]
end

subgraph group_capabilities["Assistant Capabilities"]
  node_skills["Voice Skills<br/>[skills.py]"]
  node_app_discovery["App Discovery<br/>[appindex.py]"]
  node_answers["Offline Answers<br/>[answers.py]"]
  node_ai["Claude Brain<br/>[brain.py]"]
  node_reminders[("Timers and Reminders<br/>[reminders.py]")]
  node_gesture["Gesture Controls<br/>[skills.py]"]
end

subgraph group_identity["Identity and Personalization"]
  node_voice_auth["Voice Verification<br/>[voiceauth.py]"]
  node_voice_enroll["Voice Enrollment<br/>[enroll_voice.py]"]
  node_face_auth["Face Recognition<br/>[faceauth.py]"]
  node_face_enroll["Face Enrollment<br/>[enroll_face.py]"]
  node_proactive["Proactive Speech<br/>[proactive.py]"]
end

subgraph group_experience["Visual Experience"]
  node_hud["Live HUD<br/>[hud.py]"]
  node_boot["Boot Sequence<br/>[boot.py]"]
end

subgraph group_shared["Shared Runtime"]
  node_paths["Path Resolution<br/>[paths.py]"]
end

node_user(("User"))
node_claude{{"Claude API"}}
node_web_answers{{"Web Answer APIs"}}

node_user -->|"speaks commands"| node_app
node_app -->|"captures and speaks"| node_speech
node_app -->|"dispatches speech"| node_routing
node_app -->|"checks speaker"| node_voice_auth
node_app -->|"uses recognition"| node_face_auth
node_app -->|"starts reminders"| node_reminders
node_app -->|"starts scheduler"| node_proactive
node_app -->|"opens HUD"| node_hud
node_app -->|"runs boot"| node_boot
node_routing -->|"dispatches commands"| node_skills
node_routing -->|"falls back"| node_ai
node_routing -->|"uses offline answers"| node_answers
node_skills -->|"resolves apps"| node_app_discovery
node_skills -->|"looks up questions"| node_answers
node_skills -->|"schedules timers"| node_reminders
node_skills -->|"launches by voice"| node_gesture
node_ai -.->|"sends prompts"| node_claude
node_answers -->|"queries APIs"| node_web_answers
node_voice_enroll -->|"enrolls voiceprint"| node_voice_auth
node_face_enroll -->|"trains face model"| node_face_auth
node_app -->|"shares live state"| node_speech
node_hud -->|"reads voice state"| node_speech
node_app_discovery -->|"stores app cache"| node_paths
node_answers -->|"stores question log"| node_paths
node_voice_auth -->|"stores voiceprint"| node_paths
node_face_auth -->|"stores face data"| node_paths
node_reminders -->|"resolves storage"| node_paths

click node_app "https://github.com/gauranshahuja/jarvis/blob/main/jarvis.py"
click node_speech "https://github.com/gauranshahuja/jarvis/blob/main/core.py"
click node_routing "https://github.com/gauranshahuja/jarvis/blob/main/brain.py"
click node_skills "https://github.com/gauranshahuja/jarvis/blob/main/skills.py"
click node_app_discovery "https://github.com/gauranshahuja/jarvis/blob/main/appindex.py"
click node_answers "https://github.com/gauranshahuja/jarvis/blob/main/answers.py"
click node_ai "https://github.com/gauranshahuja/jarvis/blob/main/brain.py"
click node_reminders "https://github.com/gauranshahuja/jarvis/blob/main/reminders.py"
click node_gesture "https://github.com/gauranshahuja/jarvis/blob/main/skills.py"
click node_voice_auth "https://github.com/gauranshahuja/jarvis/blob/main/voiceauth.py"
click node_voice_enroll "https://github.com/gauranshahuja/jarvis/blob/main/enroll_voice.py"
click node_face_auth "https://github.com/gauranshahuja/jarvis/blob/main/faceauth.py"
click node_face_enroll "https://github.com/gauranshahuja/jarvis/blob/main/enroll_face.py"
click node_proactive "https://github.com/gauranshahuja/jarvis/blob/main/proactive.py"
click node_hud "https://github.com/gauranshahuja/jarvis/blob/main/hud.py"
click node_boot "https://github.com/gauranshahuja/jarvis/blob/main/boot.py"
click node_paths "https://github.com/gauranshahuja/jarvis/blob/main/paths.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_app,node_speech,node_routing,node_user toneBlue
class node_skills,node_app_discovery,node_answers,node_ai,node_reminders,node_gesture toneAmber
class node_voice_auth,node_voice_enroll,node_face_auth,node_face_enroll,node_proactive,node_claude,node_web_answers toneMint
class node_hud,node_boot toneRose
class node_paths toneIndigo
```

## Components

### Voice Interaction

- [Assistant App](https://github.com/gauranshahuja/jarvis/blob/main/jarvis.py): `jarvis.py`
- [Speech I/O](https://github.com/gauranshahuja/jarvis/blob/main/core.py): `core.py`
- [Command Brain](https://github.com/gauranshahuja/jarvis/blob/main/brain.py): `brain.py`

### Assistant Capabilities

- [Voice Skills](https://github.com/gauranshahuja/jarvis/blob/main/skills.py): `skills.py`
- [App Discovery](https://github.com/gauranshahuja/jarvis/blob/main/appindex.py): `appindex.py`
- [Offline Answers](https://github.com/gauranshahuja/jarvis/blob/main/answers.py): `answers.py`
- [Claude Brain](https://github.com/gauranshahuja/jarvis/blob/main/brain.py): `brain.py`
- [Timers and Reminders](https://github.com/gauranshahuja/jarvis/blob/main/reminders.py): `reminders.py`
- [Gesture Controls](https://github.com/gauranshahuja/jarvis/blob/main/skills.py): `skills.py`

### Identity and Personalization

- [Voice Verification](https://github.com/gauranshahuja/jarvis/blob/main/voiceauth.py): `voiceauth.py`
- [Voice Enrollment](https://github.com/gauranshahuja/jarvis/blob/main/enroll_voice.py): `enroll_voice.py`
- [Face Recognition](https://github.com/gauranshahuja/jarvis/blob/main/faceauth.py): `faceauth.py`
- [Face Enrollment](https://github.com/gauranshahuja/jarvis/blob/main/enroll_face.py): `enroll_face.py`
- [Proactive Speech](https://github.com/gauranshahuja/jarvis/blob/main/proactive.py): `proactive.py`

### Visual Experience

- [Live HUD](https://github.com/gauranshahuja/jarvis/blob/main/hud.py): `hud.py`
- [Boot Sequence](https://github.com/gauranshahuja/jarvis/blob/main/boot.py): `boot.py`

### Shared Runtime

- [Path Resolution](https://github.com/gauranshahuja/jarvis/blob/main/paths.py): `paths.py`

### Other components

- **User**
- **Claude API**
- **Web Answer APIs**

## Connections

- User → Assistant App: speaks commands (evidence: `README.md`)
- Assistant App → Speech I/O: captures and speaks (evidence: `README.md`)
- Assistant App → Command Brain: dispatches speech (evidence: `jarvis.py`)
- Assistant App → Voice Verification: checks speaker (evidence: `jarvis.py`)
- Assistant App → Face Recognition: uses recognition (evidence: `jarvis.py`)
- Assistant App → Timers and Reminders: starts reminders (evidence: `jarvis.py`)
- Assistant App → Proactive Speech: starts scheduler (evidence: `jarvis.py`)
- Assistant App → Live HUD: opens HUD (evidence: `jarvis.py`)
- Assistant App → Boot Sequence: runs boot (evidence: `jarvis.py`)
- Command Brain → Voice Skills: dispatches commands (evidence: `brain.py`)
- Command Brain → Claude Brain: falls back (evidence: `brain.py`)
- Command Brain → Offline Answers: uses offline answers (evidence: `brain.py`)
- Voice Skills → App Discovery: resolves apps (evidence: `skills.py`)
- Voice Skills → Offline Answers: looks up questions (evidence: `skills.py`)
- Voice Skills → Timers and Reminders: schedules timers (evidence: `skills.py`)
- Voice Skills → Gesture Controls: launches by voice (evidence: `README.md`)
- Claude Brain → Claude API: sends prompts (evidence: `brain.py`)
- Offline Answers → Web Answer APIs: queries APIs (evidence: `answers.py`)
- Voice Enrollment → Voice Verification: enrolls voiceprint (evidence: `enroll_voice.py`)
- Face Enrollment → Face Recognition: trains face model (evidence: `enroll_face.py`)
- Assistant App → Speech I/O: shares live state (evidence: `jarvis.py`)
- Live HUD → Speech I/O: reads voice state (evidence: `hud.py`)
- App Discovery → Path Resolution: stores app cache (evidence: `appindex.py`)
- Offline Answers → Path Resolution: stores question log (evidence: `answers.py`)
- Voice Verification → Path Resolution: stores voiceprint (evidence: `voiceauth.py`)
- Face Recognition → Path Resolution: stores face data (evidence: `faceauth.py`)
- Timers and Reminders → Path Resolution: resolves storage (evidence: `reminders.py`)
