# samadritaacharya/meeting-intelligence-agent architecture

> Architecture diagram of the GitHub repository samadritaacharya/meeting-intelligence-agent, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/samadritaacharya/meeting-intelligence-agent
- Repository: https://github.com/samadritaacharya/meeting-intelligence-agent

## Overview

Meeting Intelligence Agent is a Streamlit application that turns pasted transcripts, uploaded audio, or sample content into a structured meeting decision package, then presents it and offers downloadable reports and operational artifacts. Analysis can use Anthropic or a deterministic fallback; local Whisper transcription is optional. The excerpts show the main analysis call and PDF export call, while the README documents the broader export formats and product outputs. Some app wiring for audio and non-PDF exports is not excerpted, so those links are marked unverified.

Main entry points: [`app.py`](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py) (Streamlit workspace), [`utils/analyzer.py`](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py) (Meeting analyzer), [`utils/exporter.py`](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py) (PDF report)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_experience["Meeting workspace"]
  node_app["Streamlit workspace<br/>[app.py]"]
  node_inputs["Transcript and audio input<br/>[app.py]"]
  node_decision_package["Decision package views<br/>[app.py]"]
end

subgraph group_analysis["Analysis pipeline"]
  node_analyzer["Meeting analyzer<br/>[analyzer.py]"]
  node_fallback["Deterministic fallback<br/>[analyzer.py]"]
  node_anthropic["Anthropic analysis<br/>[analyzer.py]"]
  node_whisper["Whisper transcription<br/>[analyzer.py]"]
  node_normalization["Response normalization<br/>[analyzer.py]"]
end

subgraph group_artifacts["Operational artifacts"]
  node_pdf["PDF report<br/>[exporter.py]"]
  node_docx["DOCX report<br/>[exporter.py]"]
  node_jira_csv["Jira action CSV<br/>[exporter.py]"]
  node_risk_csv["Risk register CSV<br/>[exporter.py]"]
  node_markdown["Markdown notes<br/>[exporter.py]"]
end

node_user(("Meeting user"))
node_anthropic_service{{"Anthropic API"}}
node_whisper_runtime{{"Local Whisper"}}

node_user -->|"uses"| node_app
node_app -->|"collects"| node_inputs
node_app -->|"analyzes transcript"| node_analyzer
node_inputs -->|"provides meeting content"| node_app
node_app -->|"presents results"| node_decision_package
node_analyzer -->|"falls back to"| node_fallback
node_analyzer -->|"requests analysis"| node_anthropic
node_anthropic -->|"calls"| node_anthropic_service
node_analyzer -->|"normalizes response"| node_normalization
node_app -.->|"transcribes audio"| node_whisper
node_whisper -->|"uses"| node_whisper_runtime
node_app -->|"exports PDF"| node_pdf
node_app -.->|"offers DOCX download"| node_docx
node_app -.->|"offers action CSV"| node_jira_csv
node_app -.->|"offers risk CSV"| node_risk_csv
node_app -.->|"offers Markdown"| node_markdown

click node_app "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py"
click node_inputs "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py"
click node_decision_package "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py"
click node_analyzer "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py"
click node_fallback "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py"
click node_anthropic "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py"
click node_whisper "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py"
click node_normalization "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py"
click node_pdf "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py"
click node_docx "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py"
click node_jira_csv "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py"
click node_risk_csv "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py"
click node_markdown "https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.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_inputs,node_decision_package,node_user toneBlue
class node_analyzer,node_fallback,node_anthropic,node_whisper,node_normalization toneAmber
class node_pdf,node_docx,node_jira_csv,node_risk_csv,node_markdown,node_anthropic_service toneMint
class node_whisper_runtime toneIndigo
```

## Components

### Meeting workspace

- [Streamlit workspace](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py): `app.py`
- [Transcript and audio input](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py): `app.py`
- [Decision package views](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/app.py): `app.py`

### Analysis pipeline

- [Meeting analyzer](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py): `utils/analyzer.py`
- [Deterministic fallback](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py): `utils/analyzer.py`
- [Anthropic analysis](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py): `utils/analyzer.py`
- [Whisper transcription](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py): `utils/analyzer.py`
- [Response normalization](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/analyzer.py): `utils/analyzer.py`

### Operational artifacts

- [PDF report](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py): `utils/exporter.py`
- [DOCX report](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py): `utils/exporter.py`
- [Jira action CSV](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py): `utils/exporter.py`
- [Risk register CSV](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py): `utils/exporter.py`
- [Markdown notes](https://github.com/samadritaacharya/meeting-intelligence-agent/blob/main/utils/exporter.py): `utils/exporter.py`

### Other components

- **Meeting user**
- **Anthropic API**
- **Local Whisper**

## Connections

- Meeting user → Streamlit workspace: uses (evidence: `README.md`)
- Streamlit workspace → Transcript and audio input: collects (evidence: `app.py`)
- Streamlit workspace → Meeting analyzer: analyzes transcript (evidence: `app.py`)
- Transcript and audio input → Streamlit workspace: provides meeting content (evidence: `README.md`)
- Streamlit workspace → Decision package views: presents results (evidence: `app.py`)
- Meeting analyzer → Deterministic fallback: falls back to (evidence: `utils/analyzer.py`)
- Meeting analyzer → Anthropic analysis: requests analysis (evidence: `utils/analyzer.py`)
- Anthropic analysis → Anthropic API: calls (evidence: `utils/analyzer.py`)
- Meeting analyzer → Response normalization: normalizes response (evidence: `utils/analyzer.py`)
- Streamlit workspace → Whisper transcription: transcribes audio (evidence: `app.py`)
- Whisper transcription → Local Whisper: uses (evidence: `utils/analyzer.py`)
- Streamlit workspace → PDF report: exports PDF (evidence: `app.py`)
- Streamlit workspace → DOCX report: offers DOCX download (evidence: `README.md`)
- Streamlit workspace → Jira action CSV: offers action CSV (evidence: `README.md`)
- Streamlit workspace → Risk register CSV: offers risk CSV (evidence: `README.md`)
- Streamlit workspace → Markdown notes: offers Markdown (evidence: `README.md`)
