cowork-dash 0.1.4__tar.gz → 0.1.6__tar.gz

{cowork_dash-0.1.4 → cowork_dash-0.1.6}/CHANGELOG.md

@@ -5,6 +5,33 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.1.6] - 2026-01-25
+
+### Fixed
+- Duplicate agent response rendering issue
+- Agent state now resets on page refresh for clean sessions
+- Tool calls indicator disappearing during execution (race condition fix)
+- Missing return value in interrupt handling callback
+- Default agent workspace now uses environment variable or current directory
+
+## [0.1.5] - 2026-01-24
+
+### Added
+- Stop button to halt agent execution mid-run
+- Clear canvas confirmation modal with archive functionality
+- Canvas item collapse/expand and delete features
+- Folder selection and creation in file browser
+- Double-click folders to change working directory
+- Demo video and screenshots in README
+
+### Changed
+- Default agent workspace root changed to "/" for full virtual filesystem access
+- Simplified README with concise installation instructions
+- Mermaid diagrams now respect light/dark theme changes
+
+### Removed
+- requirements.txt (dependencies managed via pyproject.toml)
+
 ## [0.1.4] - 2026-01-20
 
 ### Added
@@ -91,6 +118,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Resizable split-pane interface
 - Upload/download functionality for files
 
+[0.1.6]: https://github.com/dkedar7/cowork-dash/compare/v0.1.5...v0.1.6
+[0.1.5]: https://github.com/dkedar7/cowork-dash/compare/v0.1.4...v0.1.5
 [0.1.4]: https://github.com/dkedar7/cowork-dash/compare/v0.1.3...v0.1.4
 [0.1.3]: https://github.com/dkedar7/cowork-dash/compare/v0.1.2...v0.1.3
 [0.1.2]: https://github.com/dkedar7/cowork-dash/compare/v0.1.1...v0.1.2

{cowork_dash-0.1.4 → cowork_dash-0.1.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cowork-dash
-Version: 0.1.4
+Version: 0.1.6
 Summary: AI Agent Web Interface with Filesystem and Canvas Visualization
 Project-URL: Homepage, https://github.com/dkedar7/cowork-dash
 Project-URL: Documentation, https://github.com/dkedar7/cowork-dash/blob/main/README.md
@@ -51,6 +51,11 @@ Description-Content-Type: text/markdown
 
 A web interface for AI agent interactions with filesystem workspace, canvas visualization, and real-time streaming.
 
+<p align="center">
+  <img src="docs/light.png" alt="Cowork Dash - Light Mode" width="48%">
+  <img src="docs/dark.png" alt="Cowork Dash - Dark Mode" width="48%">
+</p>
+
 ## Features
 
 - **AI Agent Chat**: Real-time streaming with thinking process and task progress
@@ -58,6 +63,15 @@ A web interface for AI agent interactions with filesystem workspace, canvas visu
 - **Canvas**: Visualize DataFrames, Plotly/Matplotlib charts, Mermaid diagrams, images
 - **Flexible Configuration**: Environment variables, CLI args, or config file
 
+## Demo
+
+<p align="center">
+  <em>Organizing files with an AI agent</em><br><br>
+  <a href="https://www.youtube.com/watch?v=yKB9cBx80VE">
+    <img src="https://img.youtube.com/vi/yKB9cBx80VE/maxresdefault.jpg" alt="Cowork Dash Demo Video" width="70%">
+  </a>
+</p>
+
 ## Quick Start
 
 ### Installation
@@ -68,7 +82,7 @@ pip install cowork-dash
 
 # Or run directly with uvx (no installation needed)
 export ANTHROPIC_API_KEY="your_anthropic_api_key"
-uvx cowork-dash run --workspace ~/my-workspace
+cowork-dash run
 ```
 
 ### Run
@@ -100,19 +114,6 @@ Open browser to `http://localhost:8050`
 2. **Environment Variables** - `DEEPAGENT_*`
 3. **Config File** - `config.py` defaults
 
-### Environment Variables (optional)
-
-```bash
-export DEEPAGENT_SPEC=my_agent.py:agent         # Set any Langgraph agent
-export DEEPAGENT_WORKSPACE_ROOT=/path/to/workspace
-export DEEPAGENT_PORT=9000                      # optional (default: 8050)
-export DEEPAGENT_HOST=0.0.0.0                   # optional (default: localhost)
-export DEEPAGENT_DEBUG=true                     # optional (default: false)
-export DEEPAGENT_APP_TITLE="My App"             # optional
-export DEEPAGENT_APP_SUBTITLE="Subtitle"        # optional
-
-cowork-dash run
-```
 
 ### CLI Options (all optional)
 
@@ -120,12 +121,26 @@ cowork-dash run
 cowork-dash run [OPTIONS]
 
   --workspace PATH        Workspace directory (default: current directory)
-  --agent PATH:OBJECT     Agent spec (default: none, manual mode)
+  --agent PATH:OBJECT     Point to your Langgraph agent (default: default agent, manual mode)
   --port PORT            Server port (default: 8050)
   --host HOST            Server host (default: localhost)
   --debug                Enable debug mode
   --title TITLE          App title (default: "Cowork Dash")
   --subtitle TEXT        App subtitle (default: "AI-Powered Workspace")
+  --welcome-message TEXT Welcome message shown in chat (supports markdown)
+```
+
+### Environment Variables (optional)
+
+```bash
+export DEEPAGENT_SPEC=my_agent.py:agent         # Set any Langgraph agent
+export DEEPAGENT_WORKSPACE_ROOT=/path/to/workspace
+export DEEPAGENT_DEBUG=true                     # optional (default: false)
+export DEEPAGENT_APP_TITLE="My App"             # optional
+export DEEPAGENT_APP_SUBTITLE="Subtitle"        # optional
+export DEEPAGENT_WELCOME_MESSAGE="Hello!"       # optional (supports markdown)
+
+cowork-dash run
 ```
 
 ### Python API
@@ -141,27 +156,10 @@ run_app(agent, workspace="~/my-workspace")
 # Option 2: Use agent spec
 run_app(agent_spec="my_agent.py:agent", workspace="~/my-workspace")
 
-# Option 3: Manual mode (no agent)
-run_app(workspace="~/my-workspace", port=8080, debug=True)
 ```
 
 ## Agent Integration
 
-### Workspace Access
-
-Cowork Dash sets `DEEPAGENT_WORKSPACE_ROOT` environment variable for your agent:
-
-```python
-import os
-from pathlib import Path
-
-# In your agent code
-workspace = Path(os.getenv('DEEPAGENT_WORKSPACE_ROOT', './'))
-
-# Read/write files in workspace
-config_file = workspace / "config.json"
-```
-
 ### Agent Specification
 
 Load agents using `path:object` format:
@@ -174,13 +172,6 @@ cowork-dash run --agent agent.py:my_agent
 cowork-dash run --agent /path/to/agent.py:agent_instance
 ```
 
-### Agent Requirements
-
-Your agent must implement:
-- **Streaming**: `agent.stream(input, stream_mode="updates")`
-- **Message format**: `{"messages": [{"role": "user", "content": "..."}]}`
-- **Workspace access** (optional): Read `DEEPAGENT_WORKSPACE_ROOT` env var
-
 ### Example Agent Setup
 
 ```python

{cowork_dash-0.1.4 → cowork_dash-0.1.6}/README.md

@@ -2,6 +2,11 @@
 
 A web interface for AI agent interactions with filesystem workspace, canvas visualization, and real-time streaming.
 
+<p align="center">
+  <img src="docs/light.png" alt="Cowork Dash - Light Mode" width="48%">
+  <img src="docs/dark.png" alt="Cowork Dash - Dark Mode" width="48%">
+</p>
+
 ## Features
 
 - **AI Agent Chat**: Real-time streaming with thinking process and task progress
@@ -9,6 +14,15 @@ A web interface for AI agent interactions with filesystem workspace, canvas visu
 - **Canvas**: Visualize DataFrames, Plotly/Matplotlib charts, Mermaid diagrams, images
 - **Flexible Configuration**: Environment variables, CLI args, or config file
 
+## Demo
+
+<p align="center">
+  <em>Organizing files with an AI agent</em><br><br>
+  <a href="https://www.youtube.com/watch?v=yKB9cBx80VE">
+    <img src="https://img.youtube.com/vi/yKB9cBx80VE/maxresdefault.jpg" alt="Cowork Dash Demo Video" width="70%">
+  </a>
+</p>
+
 ## Quick Start
 
 ### Installation
@@ -19,7 +33,7 @@ pip install cowork-dash
 
 # Or run directly with uvx (no installation needed)
 export ANTHROPIC_API_KEY="your_anthropic_api_key"
-uvx cowork-dash run --workspace ~/my-workspace
+cowork-dash run
 ```
 
 ### Run
@@ -51,19 +65,6 @@ Open browser to `http://localhost:8050`
 2. **Environment Variables** - `DEEPAGENT_*`
 3. **Config File** - `config.py` defaults
 
-### Environment Variables (optional)
-
-```bash
-export DEEPAGENT_SPEC=my_agent.py:agent         # Set any Langgraph agent
-export DEEPAGENT_WORKSPACE_ROOT=/path/to/workspace
-export DEEPAGENT_PORT=9000                      # optional (default: 8050)
-export DEEPAGENT_HOST=0.0.0.0                   # optional (default: localhost)
-export DEEPAGENT_DEBUG=true                     # optional (default: false)
-export DEEPAGENT_APP_TITLE="My App"             # optional
-export DEEPAGENT_APP_SUBTITLE="Subtitle"        # optional
-
-cowork-dash run
-```
 
 ### CLI Options (all optional)
 
@@ -71,12 +72,26 @@ cowork-dash run
 cowork-dash run [OPTIONS]
 
   --workspace PATH        Workspace directory (default: current directory)
-  --agent PATH:OBJECT     Agent spec (default: none, manual mode)
+  --agent PATH:OBJECT     Point to your Langgraph agent (default: default agent, manual mode)
   --port PORT            Server port (default: 8050)
   --host HOST            Server host (default: localhost)
   --debug                Enable debug mode
   --title TITLE          App title (default: "Cowork Dash")
   --subtitle TEXT        App subtitle (default: "AI-Powered Workspace")
+  --welcome-message TEXT Welcome message shown in chat (supports markdown)
+```
+
+### Environment Variables (optional)
+
+```bash
+export DEEPAGENT_SPEC=my_agent.py:agent         # Set any Langgraph agent
+export DEEPAGENT_WORKSPACE_ROOT=/path/to/workspace
+export DEEPAGENT_DEBUG=true                     # optional (default: false)
+export DEEPAGENT_APP_TITLE="My App"             # optional
+export DEEPAGENT_APP_SUBTITLE="Subtitle"        # optional
+export DEEPAGENT_WELCOME_MESSAGE="Hello!"       # optional (supports markdown)
+
+cowork-dash run
 ```
 
 ### Python API
@@ -92,27 +107,10 @@ run_app(agent, workspace="~/my-workspace")
 # Option 2: Use agent spec
 run_app(agent_spec="my_agent.py:agent", workspace="~/my-workspace")
 
-# Option 3: Manual mode (no agent)
-run_app(workspace="~/my-workspace", port=8080, debug=True)
 ```
 
 ## Agent Integration
 
-### Workspace Access
-
-Cowork Dash sets `DEEPAGENT_WORKSPACE_ROOT` environment variable for your agent:
-
-```python
-import os
-from pathlib import Path
-
-# In your agent code
-workspace = Path(os.getenv('DEEPAGENT_WORKSPACE_ROOT', './'))
-
-# Read/write files in workspace
-config_file = workspace / "config.json"
-```
-
 ### Agent Specification
 
 Load agents using `path:object` format:
@@ -125,13 +123,6 @@ cowork-dash run --agent agent.py:my_agent
 cowork-dash run --agent /path/to/agent.py:agent_instance
 ```
 
-### Agent Requirements
-
-Your agent must implement:
-- **Streaming**: `agent.stream(input, stream_mode="updates")`
-- **Message format**: `{"messages": [{"role": "user", "content": "..."}]}`
-- **Workspace access** (optional): Read `DEEPAGENT_WORKSPACE_ROOT` env var
-
 ### Example Agent Setup
 
 ```python

{cowork_dash-0.1.4 → cowork_dash-0.1.6}/cowork_dash/agent.py

@@ -1,3 +1,4 @@
+import os
 import uuid
 from deepagents import create_deep_agent
 from langgraph.checkpoint.memory import InMemorySaver
@@ -85,14 +86,16 @@ Work iteratively like a human using Jupyter:
 4. Keep cells focused on single tasks for easier debugging
 
 ### General
-1. Use write_todos to track your progress and next steps
-2. Use think_tool to reason through complex problems
+1. ALWAYS use write_todos to track your progress and next steps
+2. ALWAYS think_tool to reason through reqests, irrespective of complexity
 3. Be proactive in exploring the filesystem when relevant
 4. Provide clear, helpful responses
 
 The workspace is your sandbox - feel free to create files, organize content, and help users manage their projects."""
 
-backend = FilesystemBackend(root_dir=str("./"), virtual_mode=True)
+# Get workspace root from environment variable or default to current directory
+workspace_root = os.getenv("DEEPAGENT_WORKSPACE_ROOT", os.getcwd())
+backend = FilesystemBackend(root_dir=workspace_root, virtual_mode=True)
 
 agent = create_deep_agent(
     system_prompt=SYSTEM_PROMPT,
@@ -113,6 +116,6 @@ agent = create_deep_agent(
         get_notebook_canvas_items,
         clear_notebook_canvas_items,
     ],
-    # interrupt_on=dict(bash=True),
+    interrupt_on=dict(bash=True),
     checkpointer=InMemorySaver()
 )