How dinesh-dev1628/personal-agent works
Personal Agent is a FastAPI application with three user-facing workflows: habit planning, coding assistance, and general chat. Each route delegates to a matching agent runner, which uses task-specific instructions and a typed response model before returning the result. Runtime configuration supplies the Groq model and server settings. The README provides little architectural detail, so this map relies on the shown source; the model-provider interaction is inferred from the configured Groq model and agent execution.
Main entry points: brain.py Agent runners · app.py HTTP routes · index.py API adapter · instructions.py Task instructions
Components 12
API
- API adapter
api/index.py - HTTP routes
src/app.py
Agent workspaces
- Agent runners
src/ai/brain.py - Habit agent
src/ai/brain.py - Coding agent
src/ai/brain.py - General agent
src/ai/brain.py
Shared agent setup
- Task instructions
src/data/instructions.py - Response schemas
src/models/basemodel.py - Model selection
src/utils/settings.py - Runtime settings
src/utils/env_data.py
Other components
- User
- Groq model
Connections 14
- User → HTTP routes: submits prompt
src/app.py - API adapter → HTTP routes: imports app
api/index.py - HTTP routes → Agent runners: dispatches task
src/app.py - Agent runners → Habit agent: runs habit
src/ai/brain.py - Agent runners → Coding agent: runs coding
src/ai/brain.py - Agent runners → General agent: runs chat
src/ai/brain.py - Agent runners → Task instructions: configures agents
src/ai/brain.py - Agent runners → Response schemas: sets output types
src/ai/brain.py - Agent runners → Model selection: selects model
src/ai/brain.py - HTTP routes → Runtime settings: reads server settings
src/app.py - Habit agent → Groq model: requests generation
- Coding agent → Groq model: requests generation
- General agent → Groq model: requests generation
- HTTP routes → User: returns result
src/app.py