Architecture Overview

Who This Is For

• Developers extending diagnostics/remediation modules.
• Operators debugging interactions between agent and API.

High-Level Components

• agent/: PowerShell orchestration, modules, telemetry, SQLite writes
• agent/core/report.ps1: HTML, JSON, Markdown, and optional PDF report generation
• server/api/: FastAPI app, SQLite query layer, service wrapper
• server/dashboard/: static dashboard + Chart.js polling client
• config/: INI template and runtime config
• installer/: build and Inno packaging scripts

Execution Flow

1. Agent entrypoint (AutoDoctor.ps1) initializes paths and DB.
2. Module engine executes collection modules, then validation/history/anomaly/correlation.
3. Root cause and remediation run, then results are persisted.
4. HTML, JSON, Markdown, and optional PDF reports are generated.
5. Telemetry JSON and latest_run.json are written.
6. API reads SQLite + metadata + report JSON and serves endpoints.
7. Dashboard polls API every 5 seconds.

AutoDoctor.ps1
  -> module engine
  -> rootcause + remediation
  -> SQLite writes + telemetry snapshots
  -> report generation (HTML / JSON / MD / PDF)
  -> latest_run.json
  -> FastAPI endpoints
  -> dashboard UI

Service Runtime Flow

• Service wrapper (autodoctor_service.py) resolves runtime paths and mode.
• Child API process starts from bundled binary or source script depending on mode.
• API binds host/port based on precedence (registry -> INI -> env -> defaults).

Key Design Characteristics

• Local-first storage and runtime
• SQLite as shared state between writer (agent) and reader (API)
• Report JSON as a second read model for higher-level dashboard summaries
• Configurable service registration mode for constrained environments
• Clear separation of collection (agent) and presentation (API/dashboard)

Next Steps

• Read Module Engine and Data Contracts
• Read Telemetry and Persistence