Todo
Manage todo lists and generate tasks from agent responses
The todo command and --todo flag help you manage task lists and automatically generate todos from agent responses.
Quick Start
# List all todos
praisonai todo list
# Generate todo from task
praisonai "Plan the project" --todo
# Add a todo manually
praisonai todo add "Implement feature X"!Manage Task Lists and Generate
Commands
Add Todo
praisonai todo add "Implement user authentication"
# Set a priority (as of PraisonAI PR #4946, -p/--priority is honoured)
praisonai todo add "Ship the release" --priority high
praisonai todo add "Tidy up docs" -p lowExpected Output:
✅ Added: Implement user authentication
📋 Todo List:
--------------------------------------------------
1. ⬜ 🟡 Implement user authentication
--------------------------------------------------
0/1 completed| Flag | Short | Values | Default | Description |
|---|---|---|---|---|
--priority | -p | low, medium, high | medium | Priority for the new todo. high → 🔴, medium → 🟡, low → 🟢. Unknown values are rejected. |
List Todos
praisonai todo listExpected Output:
📋 Todo List:
--------------------------------------------------
1. ⬜ 🔴 Fix critical bug in payment system
2. ⬜ 🟡 Implement user authentication
3. ⬜ 🟡 Write API documentation
4. ✅ 🟢 Set up CI/CD pipeline
5. ⬜ 🟢 Add unit tests
--------------------------------------------------
1/5 completed
Legend: 🔴 High 🟡 Medium 🟢 Low
⬜ Pending ✅ CompletedComplete Todo
praisonai todo complete 2Expected Output:
✅ Completed: Implement user authentication
📋 Todo List:
--------------------------------------------------
1. ⬜ 🔴 Fix critical bug in payment system
2. ✅ 🟡 Implement user authentication
3. ⬜ 🟡 Write API documentation
4. ✅ 🟢 Set up CI/CD pipeline
5. ⬜ 🟢 Add unit tests
--------------------------------------------------
2/5 completedDelete Todo
praisonai todo delete 3Expected Output:
🗑️ Deleted: Write API documentation
📋 Todo List:
--------------------------------------------------
1. ⬜ 🔴 Fix critical bug in payment system
2. ✅ 🟡 Implement user authentication
3. ✅ 🟢 Set up CI/CD pipeline
4. ⬜ 🟢 Add unit tests
--------------------------------------------------
2/4 completedClear Todos
praisonai todo clearExpected Output:
⚠️ Clear all todos? (y/N): y
🗑️ All todos cleared.Help
praisonai todo helpExpected Output:
Todo Commands:
praisonai todo add <task> - Add a new todo item
praisonai todo list - List all todos
praisonai todo complete <id> - Mark todo as complete
praisonai todo delete <id> - Delete a todo
praisonai todo clear - Clear all todos
praisonai todo help - Show this help
Using --todo Flag:
praisonai "task" --todo - Generate todos from responseUsing –todo Flag
Generate Todos from Task
praisonai "Plan the implementation of a REST API" --todoExpected Output:
╭─ Agent Info ─────────────────────────────────────────────────────────────────╮
│ 👤 Agent: DirectAgent │
│ Role: Assistant │
│ 📋 Todo Generation: Enabled │
╰──────────────────────────────────────────────────────────────────────────────╯
╭────────────────────────────────── Response ──────────────────────────────────╮
│ Here's a plan for implementing the REST API: │
│ │
│ 1. Design the API schema and endpoints │
│ 2. Set up the project structure │
│ 3. Implement authentication middleware │
│ 4. Create CRUD endpoints for resources │
│ 5. Add input validation │
│ 6. Implement error handling │
│ 7. Write unit tests │
│ 8. Add API documentation │
╰──────────────────────────────────────────────────────────────────────────────╯
📋 Generated Todos:
--------------------------------------------------
1. ⬜ 🟡 Design the API schema and endpoints
2. ⬜ 🟡 Set up the project structure
3. ⬜ 🟡 Implement authentication middleware
4. ⬜ 🟡 Create CRUD endpoints for resources
5. ⬜ 🟡 Add input validation
6. ⬜ 🟡 Implement error handling
7. ⬜ 🟢 Write unit tests
8. ⬜ 🟢 Add API documentation
--------------------------------------------------
0/8 completed
✅ 8 todos generated from responseCombine with Planning
praisonai "Plan a machine learning project" --todo --planningExpected Output:
📋 PLANNING PHASE
Creating implementation plan...
[Planning output...]
📋 Generated Todos from Plan:
--------------------------------------------------
1. ⬜ 🔴 Define problem statement and success metrics
2. ⬜ 🔴 Collect and prepare training data
3. ⬜ 🟡 Perform exploratory data analysis
4. ⬜ 🟡 Feature engineering
5. ⬜ 🟡 Model selection and training
6. ⬜ 🟡 Hyperparameter tuning
7. ⬜ 🟢 Model evaluation
8. ⬜ 🟢 Deploy model to production
--------------------------------------------------
0/8 completedUse Cases
Project Planning
praisonai "Plan the development of a mobile app" --todoSprint Planning
praisonai "Break down the user story: As a user, I want to reset my password" --todoLearning Path
praisonai "Create a learning plan for Kubernetes" --todoBug Triage
praisonai "Analyze this error and suggest fixes" --todo --fast-context ./srcTodo Storage
Todos are stored in the home-level workspace ~/.praison/todos.json:
{
"todos": [
{
"id": 1,
"task": "Implement user authentication",
"priority": "medium",
"completed": false,
"created": "2024-12-16T15:30:00Z"
}
]
}Priority Levels
| Priority | Icon | Description |
|---|---|---|
| High | 🔴 | Critical tasks, blockers |
| Medium | 🟡 | Standard priority |
| Low | 🟢 | Nice-to-have, can wait |
Best Practices
Use praisonai todo list regularly to track progress
Periodically clear completed todos to keep list manageable
Combine with PlanningUse --planning --todo for detailed task breakdowns
Consider using different workspaces for different projects
Workflow Example
# 1. Generate todos from project plan
praisonai "Plan the API implementation" --todo --planning
# 2. View todos
praisonai todo list
# 3. Work on first task
praisonai "Help me design the API schema" --fast-context ./src
# 4. Mark as complete
praisonai todo complete 1
# 5. Continue with next task
praisonai todo listVia MCP
The praisonai-mcp host exposes four MCP tools — praisonai.todo.list, praisonai.todo.add, praisonai.todo.complete, praisonai.todo.delete — that read and write the runtime todo store (<workspace>/todos.json, else ~/.praisonai/todos.json). A todo added from Claude Desktop is visible to an agent run on the same machine, and vice versa.
See MCP Todo Tools for the tool table, record shape, and store resolution.