Merge pull request #1336 from hughiwnl/autosavedocs

Updated docs for autosave feature

Author: kyegomez

docs/swarms/examples/concurrent_workflow.md

@@ -175,16 +175,28 @@ workflow = ConcurrentWorkflow(
 )
 ```
 
-#### Conversation History Management
+## Autosave Feature
 
-Automatic conversation tracking and persistence:
+Autosave is enabled by default (`autosave=True`). Conversation history is automatically saved to `{workspace_dir}/swarms/ConcurrentWorkflow/{workflow-name}-{timestamp}/conversation_history.json`.
+
+To set a custom workspace directory name, use the `WORKSPACE_DIR` environment variable:
 
 ```python
+import os
+from swarms import Agent, ConcurrentWorkflow
+
+# Set custom workspace directory where conversation history will be saved
+# If not set, defaults to 'agent_workspace' in the current directory
+os.environ["WORKSPACE_DIR"] = "my_project"
+
+# Create workflow (autosave enabled by default)
 workflow = ConcurrentWorkflow(
-    agents=agents,
-    auto_save=True,                    # Auto-save conversation history
-    metadata_output_path="results.json"  # Custom output file path
+    name="analysis-workflow",
+    agents=[analyst1, analyst2, analyst3],
 )
+
+# Run workflow - conversation automatically saved
+result = workflow.run("Analyze market trends")
 ```
 
 #### Multimodal Support

docs/swarms/structs/hierarchical_swarm.md

@@ -66,6 +66,8 @@ Initializes a new HierarchicalSwarm instance.
 | `multi_agent_prompt_improvements` | `bool` | `False` | No | Enable enhanced multi-agent collaboration prompts |
 | `interactive` | `bool` | `False` | No | Enable interactive mode with Hierarchical Swarms dashboard visualization |
 | `planning_enabled` | `bool` | `True` | No | Enable planning phase before task distribution |
+| `autosave` | `bool` | `True` | No | Whether to enable autosaving of conversation history |
+| `verbose` | `bool` | `False` | No | Whether to enable verbose logging |
 
 #### Returns
 
@@ -481,6 +483,31 @@ The dashboard automatically:
 | **Detailed View** | Full output history for each agent in each loop |
 | **Real-time Updates** | Dashboard refreshes automatically as operations progress |
 
+## Autosave Feature
+
+Autosave is enabled by default (`autosave=True`). Conversation history is automatically saved to `{workspace_dir}/swarms/HierarchicalSwarm/{swarm-name}-{timestamp}/conversation_history.json` after all loops complete.
+
+To set a custom workspace directory name, use the `WORKSPACE_DIR` environment variable:
+
+```python
+import os
+from swarms import Agent, HierarchicalSwarm
+
+# Set custom workspace directory where conversation history will be saved
+# If not set, defaults to 'agent_workspace' in the current directory
+os.environ["WORKSPACE_DIR"] = "my_project"
+
+# Create swarm (autosave enabled by default)
+swarm = HierarchicalSwarm(
+    name="analysis-swarm",
+    agents=[research_agent, financial_agent],
+    max_loops=2,
+)
+
+# Run swarm - conversation automatically saved after all loops
+result = swarm.run("Analyze Tesla (TSLA) stock")
+```
+
 ## Planning Feature
 
 The `planning_enabled` parameter controls whether the director performs an initial planning phase before creating task orders. When enabled, the director:
@@ -596,6 +623,7 @@ def live_paragraph_callback(agent_name: str, chunk: str, is_final: bool):
 | **Planning Strategy**        | Enable `planning_enabled` for complex tasks, disable for simple or performance-critical tasks   |
 | **Interactive Dashboard**    | Use `interactive=True` during development for real-time monitoring and debugging                 |
 | **Context Preservation**     | Leverage the built-in conversation history for continuity                                       |
+| **Autosave Configuration**   | Keep `autosave=True` (default) to preserve conversation history for debugging and analysis     |
 | **Error Handling**           | Implement proper error handling for production use                                              |
 | **Streaming Callbacks**      | Use streaming callbacks for real-time monitoring and user feedback                              |
 | **Callback Performance**     | Keep streaming callbacks lightweight to avoid blocking the main execution thread                |

docs/swarms/structs/sequential_workflow.md

@@ -34,8 +34,11 @@ graph TD
 | `shared_memory_system` | `callable` | Optional callable for managing shared memory between agents. |
 | `multi_agent_collab_prompt` | `bool` | If True, appends a collaborative prompt to each agent's system prompt. |
 | `team_awareness` | `bool`        | Enables sequential awareness features (passed to internal `AgentRearrange`). Defaults to `False`. |
+| `autosave` | `bool` | Whether to enable autosaving of conversation history. Defaults to `True`. |
+| `verbose` | `bool` | Whether to enable verbose logging. Defaults to `False`. |
 | `flow`           | `str`         | A string representing the order of agents (e.g., "Agent1 -> Agent2 -> Agent3"). |
 | `agent_rearrange`| `AgentRearrange` | Internal helper for managing agent execution. |
+| `swarm_workspace_dir` | `str` | The workspace directory where conversation history is saved (set automatically when autosave is enabled). |
 
 ## Methods
 
@@ -53,6 +56,8 @@ The constructor initializes the `SequentialWorkflow` object.
   - `shared_memory_system` (`callable`, optional): Callable for shared memory management. Defaults to `None`.
   - `multi_agent_collab_prompt` (`bool`, optional): If True, appends a collaborative prompt to each agent's system prompt. Defaults to `False`.
   - `team_awareness` (`bool`, optional): Enables sequential awareness features in the underlying `AgentRearrange`. Defaults to `False`.
+  - `autosave` (`bool`, optional): Whether to enable autosaving of conversation history. Defaults to `True`.
+  - `verbose` (`bool`, optional): Whether to enable verbose logging. Defaults to `False`.
   - `*args`: Variable length argument list.
   - `**kwargs`: Arbitrary keyword arguments.
 
@@ -308,6 +313,32 @@ print(result)
 | `description` | Description of workflow purpose | Standard description |
 | `max_loops` | Number of times to execute workflow | 1 |
 | `team_awareness` | Enable sequential awareness features | False |
+| `autosave` | Enable automatic saving of conversation history | True |
+| `verbose` | Enable verbose logging | False |
+
+## Autosave Feature
+
+Autosave is enabled by default (`autosave=True`). Conversation history is automatically saved to `{workspace_dir}/swarms/SequentialWorkflow/{workflow-name}-{timestamp}/conversation_history.json`.
+
+To set a custom workspace directory name, use the `WORKSPACE_DIR` environment variable:
+
+```python
+import os
+from swarms import Agent, SequentialWorkflow
+
+# Set custom workspace directory where conversation history will be saved
+# If not set, defaults to 'agent_workspace' in the current directory
+os.environ["WORKSPACE_DIR"] = "my_project"
+
+# Create workflow (autosave enabled by default)
+workflow = SequentialWorkflow(
+    name="content-workflow",
+    agents=[writer, editor],
+)
+
+# Run workflow - conversation automatically saved
+result = workflow.run("Write a blog post about AI")
+```
 
 ## Best Practices