PlanOpticon

Update docs, README, and add interactive CLI menu - Replace README with accurate feature list, install instructions, and output structure - Delete stale README_new.md - Add agent-analyze and auth commands to CLI reference docs - Add periodic capture and face detection to configuration docs - Remove "Coming soon" from cloud sources (already implemented) - Update index features and copyright year - Add interactive numbered menu when running planopticon with no args

leo 2026-02-15 04:37 trunk

Commit ecf907c803bf1dec7776f6d5f27ba46f2121b5d633061cd432d471525772c52d

Parent 287a3bb0cdafafd…

8 files changed +101 -176 -149 +50 +10 -1 -3 +5 -2 +1 -1 +103 -1

~ README.md - README_new.md ~ docs/cli-reference.md ~ docs/getting-started/configuration.md ~ docs/guide/cloud-sources.md ~ docs/index.md ~ mkdocs.yml ~ video_processor/cli/commands.py

M README.md

+101 -176

		--- README.md
		+++ README.md
		@@ -1,176 +1,101 @@
1		-PlanOpticon
2		-Comprehensive Video Analysis & Knowledge Extraction CLI
3		-PlanOpticon is an advanced AI-powered CLI tool that conducts thorough analysis of video content, extracting structured knowledge, diagrams, and actionable insights. Using state-of-the-art computer vision and natural language processing techniques, PlanOpticon transforms video assets into valuable, structured information.
4		-
5		-Core Features
6		-
7		-Complete Transcription: Full speech-to-text with speaker attribution and semantic segmentation
8		-Visual Element Extraction: Automated recognition and digitization of diagrams, charts, whiteboards, and visual aids
9		-Action Item Detection: Intelligent identification and prioritization of tasks, commitments, and follow-ups
10		-Knowledge Structure: Organization of extracted content into searchable, related concepts
11		-Plan Generation: Synthesis of extracted elements into cohesive action plans and summaries
12		-
13		-
14		-Technical Implementation
15		-PlanOpticon leverages cloud APIs and efficient processing pipelines to achieve comprehensive video analysis:
16		-Architecture Overview
17		-```
18		-Video Input → Frame Extraction → Cloud API Integration → Knowledge Integration → Structured Output
19		- ↓ ↓ ↓
20		- Frame Selection API Request Management Result Processing
21		- • Key Frame • Vision API Calls • Content Organization
22		- • Scene Detection • Speech-to-Text API • Relationship Mapping
23		- • Content Changes • LLM Analysis API • Mermaid Generation
24		-```
25		-Key Components
26		-
27		-Cloud API integration for speech-to-text transcription
28		-Vision API utilization for diagram and visual content detection
29		-LLM-powered content analysis and summarization
30		-Efficient prompt engineering for specialized content extraction
31		-Knowledge integration system for relationship mapping and organization
32		-
33		-
34		-Installation
35		-bash# Clone the repository
36		-git clone https://github.com/yourusername/planopticon.git
37		-cd planopticon
38		-
39		-# Create virtual environment
40		-python -m venv venv
41		-source venv/bin/activate # On Windows: venv\Scripts\activate
42		-
43		-# Install dependencies
44		-pip install -r requirements.txt
45		-
46		-# Install optional GPU dependencies (if available)
47		-pip install -r requirements-gpu.txt
48		-
49		-Usage
50		-PlanOpticon is designed as a command-line interface tool:
51		-bash# Basic usage
52		-planopticon analyze --input video.mp4 --output analysis/
53		-
54		-# Specify processing depth
55		-planopticon analyze --input video.mp4 --depth comprehensive --output analysis/
56		-
57		-# Focus on specific extraction types
58		-planopticon analyze --input video.mp4 --focus "diagrams,action-items" --output analysis/
59		-
60		-# Process with GPU acceleration
61		-planopticon analyze --input video.mp4 --use-gpu --output analysis/
62		-Output Structure
63		-analysis/
64		-├── transcript.json # Full transcription with timestamps and speakers
65		-├── key_points.md # Extracted main concepts and ideas
66		-├── diagrams/ # Extracted and digitized visual elements
67		-│ ├── diagram_001.svg
68		-│ └── whiteboard_001.svg
69		-├── action_items.json # Prioritized tasks and commitments
70		-└── knowledge_graph.json # Relationship map of concepts
71		-
72		-Development Guidelines
73		-When contributing to PlanOpticon, please adhere to these principles:
74		-Code Standards
75		-
76		-Follow PEP 8 style guidelines for all Python code
77		-Write comprehensive docstrings using NumPy/Google style
78		-Maintain test coverage above 80%
79		-Use type hints consistently throughout the codebase
80		-
81		-Architecture Considerations
82		-
83		-Optimize for cross-platform compatibility (macOS, Linux, Windows)
84		-Ensure ARM architecture support for cloud deployment and Apple Silicon
85		-Implement graceful degradation when GPU is unavailable
86		-Design modular components with clear interfaces
87		-
88		-
89		-System Requirements
90		-
91		-Python 3.9+
92		-8GB RAM minimum (16GB recommended)
93		-2GB disk space for models and dependencies
94		-CUDA-compatible GPU (optional, for accelerated processing)
95		-ARM64 or x86_64 architecture
96		-
97		-
98		-Implementation Strategy
99		-The core processing pipeline requires thoughtful implementation of several key systems:
100		-
101		-Frame extraction and analysis
102		-
103		-Implement selective sampling based on visual change detection
104		-Utilize region proposal networks for element identification
105		-
106		-
107		-Speech processing
108		-
109		-Apply time-domain speaker diarization
110		-Implement context-aware transcription with domain adaptation
111		-
112		-
113		-Visual element extraction
114		-
115		-Develop whiteboard/diagram detection with boundary recognition
116		-Implement reconstruction of visual elements into vector formats
117		-
118		-
119		-Knowledge integration
120		-
121		-Create hierarchical structure of extracted concepts
122		-Generate relationship mappings between identified elements
123		-
124		-
125		-Action item synthesis
126		-
127		-Apply intent recognition for commitment identification
128		-Implement priority scoring based on contextual importance
129		-
130		-
131		-
132		-Each component should be implemented as a separate module with clear interfaces, allowing for independent testing and optimization.
133		-
134		-Development Approach
135		-When implementing PlanOpticon, consider these architectural principles:
136		-
137		-Pipeline Architecture
138		-
139		-Design processing stages that can operate independently
140		-Implement data passing between stages using standardized formats
141		-Enable parallelization where appropriate
142		-Consider using Python's asyncio for I/O-bound operations
143		-
144		-
145		-Performance Optimization
146		-
147		-Implement batched processing for GPU acceleration
148		-Use memory mapping for large video files
149		-Consider JIT compilation for performance-critical sections
150		-Profile and optimize bottlenecks systematically
151		-
152		-
153		-Error Handling
154		-
155		-Implement comprehensive exception handling
156		-Design graceful degradation paths for each component
157		-Provide detailed logging for troubleshooting
158		-Consider retry mechanisms for transient failures
159		-
160		-
161		-Testing Strategy
162		-
163		-Create comprehensive unit tests for each module
164		-Implement integration tests for end-to-end pipeline
165		-Develop benchmark tests for performance evaluation
166		-Use property-based testing for complex components
167		-
168		-
169		-
170		-The implementation should maintain separation of concerns while ensuring efficient data flow between components. Consider using dependency injection patterns to improve testability and component isolation.
171		-
172		-License
173		-MIT License
174		-
175		-Contact
176		-For questions or contributions, please open an issue on GitHub or contact the maintainers at [email protected].
	1	+# PlanOpticon
	2	+
	3	+AI-powered video analysis and knowledge extraction.
	4	+
	5	+PlanOpticon processes video recordings into structured knowledge — transcripts, diagrams, action items, key points, and knowledge graphs. It auto-discovers available models across OpenAI, Anthropic, and Gemini, and produces rich multi-format output.
	6	+
	7	+## Features
	8	+
	9	+- Multi-provider AI — Auto-discovers and routes to the best available model across OpenAI, Anthropic, and Google Gemini
	10	+- Smart frame extraction — Change detection for transitions + periodic capture for slow-evolving content (document scrolling, screen shares)
	11	+- People frame filtering — OpenCV face detection automatically removes webcam/video conference frames, keeping only shared content
	12	+- Diagram extraction — Vision model classification detects flowcharts, architecture diagrams, charts, and whiteboards
	13	+- Knowledge graphs — Extracts entities and relationships, builds and merges knowledge graphs across videos
	14	+- Action item detection — Finds commitments, tasks, and follow-ups with assignees and deadlines
	15	+- Batch processing — Process entire folders of videos with merged knowledge graphs and cross-referencing
	16	+- Rich output — Markdown, HTML, PDF reports. Mermaid diagrams, SVG/PNG renderings, JSON manifests
	17	+- Cloud sources — Fetch videos from Google Drive and Dropbox shared folders
	18	+- Checkpoint/resume — Pipeline resumes from where it left off if interrupted
	19	+- Screengrab fallback — When extraction isn't perfect, captures frames with captions — something is always better than nothing
	20	+
	21	+## Quick Start
	22	+
	23	+```bash
	24	+# Install
	25	+pip install planopticon
	26	+
	27	+# Analyze a single video
	28	+planopticon analyze -i meeting.mp4 -o ./output
	29	+
	30	+# Process a folder of videos
	31	+planopticon batch -i ./recordings -o ./output --title "Weekly Meetings"
	32	+
	33	+# See available AI models
	34	+planopticon list-models
	35	+```
	36	+
	37	+## Installation
	38	+
	39	+### From PyPI
	40	+
	41	+```bash
	42	+pip install planopticon
	43	+
	44	+# With all extras (PDF, cloud sources, GPU)
	45	+pip install planopticon[all]
	46	+```
	47	+
	48	+### From Source
	49	+
	50	+```bash
	51	+git clone https://github.com/ConflictHQ/PlanOpticon.git
	52	+cd PlanOpticon
	53	+pip install -e ".[dev]"
	54	+```
	55	+
	56	+### Binary Download
	57	+
	58	+Download standalone binaries (no Python required) from [GitHub Releases](https://github.com/ConflictHQ/PlanOpticon/releases).
	59	+
	60	+### Requirements
	61	+
	62	+- Python 3.10+
	63	+- FFmpeg (`brew install ffmpeg` / `apt install ffmpeg`)
	64	+- At least one API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY`
	65	+
	66	+## Output Structure
	67	+
	68	+```
	69	+output/
	70	+├── manifest.json # Single source of truth
	71	+├── transcript/
	72	+│ ├── transcript.json # Full transcript with timestamps
	73	+│ ├── transcript.txt # Plain text
	74	+│ └── transcript.srt # Subtitles
	75	+├── frames/ # Content frames (people filtered out)
	76	+├── diagrams/ # Detected diagrams + mermaid code
	77	+├── captures/ # Screengrab fallbacks
	78	+└── results/
	79	+ ├── analysis.md # Markdown report
	80	+ ├── analysis.html # HTML report
	81	+ ├── analysis.pdf # PDF report
	82	+ ├── knowledge_graph.json # Entities and relationships
	83	+ ├── key_points.json # Extracted key points
	84	+ └── action_items.json # Tasks and follow-ups
	85	+```
	86	+
	87	+## Processing Depth
	88	+
	89	+\| Depth \| What you get \|
	90	+\|-------\|-------------\|
	91	+\| `basic` \| Transcription, key points, action items \|
	92	+\| `standard` \| + Diagram extraction (10 frames), knowledge graph, full reports \|
	93	+\| `comprehensive` \| + More frames analyzed (20), deeper extraction \|
	94	+
	95	+## Documentation
	96	+
	97	+Full documentation at [planopticon.dev](https://planopticon.dev)
	98	+
	99	+## License
	100	+
	101	+MIT License — Copyright (c) 2026 CONFLICT LLC
177	102
178	103	DELETED README_new.md

	--- README.md
	+++ README.md
	@@ -1,176 +1,101 @@
1	PlanOpticon
2	Comprehensive Video Analysis & Knowledge Extraction CLI
3	PlanOpticon is an advanced AI-powered CLI tool that conducts thorough analysis of video content, extracting structured knowledge, diagrams, and actionable insights. Using state-of-the-art computer vision and natural language processing techniques, PlanOpticon transforms video assets into valuable, structured information.
4
5	Core Features
6
7	Complete Transcription: Full speech-to-text with speaker attribution and semantic segmentation
8	Visual Element Extraction: Automated recognition and digitization of diagrams, charts, whiteboards, and visual aids
9	Action Item Detection: Intelligent identification and prioritization of tasks, commitments, and follow-ups
10	Knowledge Structure: Organization of extracted content into searchable, related concepts
11	Plan Generation: Synthesis of extracted elements into cohesive action plans and summaries
12
13
14	Technical Implementation
15	PlanOpticon leverages cloud APIs and efficient processing pipelines to achieve comprehensive video analysis:
16	Architecture Overview
17	```
18	Video Input → Frame Extraction → Cloud API Integration → Knowledge Integration → Structured Output
19	↓ ↓ ↓
20	Frame Selection API Request Management Result Processing
21	• Key Frame • Vision API Calls • Content Organization
22	• Scene Detection • Speech-to-Text API • Relationship Mapping
23	• Content Changes • LLM Analysis API • Mermaid Generation
24	```
25	Key Components
26
27	Cloud API integration for speech-to-text transcription
28	Vision API utilization for diagram and visual content detection
29	LLM-powered content analysis and summarization
30	Efficient prompt engineering for specialized content extraction
31	Knowledge integration system for relationship mapping and organization
32
33
34	Installation
35	bash# Clone the repository
36	git clone https://github.com/yourusername/planopticon.git
37	cd planopticon
38
39	# Create virtual environment
40	python -m venv venv
41	source venv/bin/activate # On Windows: venv\Scripts\activate
42
43	# Install dependencies
44	pip install -r requirements.txt
45
46	# Install optional GPU dependencies (if available)
47	pip install -r requirements-gpu.txt
48
49	Usage
50	PlanOpticon is designed as a command-line interface tool:
51	bash# Basic usage
52	planopticon analyze --input video.mp4 --output analysis/
53
54	# Specify processing depth
55	planopticon analyze --input video.mp4 --depth comprehensive --output analysis/
56
57	# Focus on specific extraction types
58	planopticon analyze --input video.mp4 --focus "diagrams,action-items" --output analysis/
59
60	# Process with GPU acceleration
61	planopticon analyze --input video.mp4 --use-gpu --output analysis/
62	Output Structure
63	analysis/
64	├── transcript.json # Full transcription with timestamps and speakers
65	├── key_points.md # Extracted main concepts and ideas
66	├── diagrams/ # Extracted and digitized visual elements
67	│ ├── diagram_001.svg
68	│ └── whiteboard_001.svg
69	├── action_items.json # Prioritized tasks and commitments
70	└── knowledge_graph.json # Relationship map of concepts
71
72	Development Guidelines
73	When contributing to PlanOpticon, please adhere to these principles:
74	Code Standards
75
76	Follow PEP 8 style guidelines for all Python code
77	Write comprehensive docstrings using NumPy/Google style
78	Maintain test coverage above 80%
79	Use type hints consistently throughout the codebase
80
81	Architecture Considerations
82
83	Optimize for cross-platform compatibility (macOS, Linux, Windows)
84	Ensure ARM architecture support for cloud deployment and Apple Silicon
85	Implement graceful degradation when GPU is unavailable
86	Design modular components with clear interfaces
87
88
89	System Requirements
90
91	Python 3.9+
92	8GB RAM minimum (16GB recommended)
93	2GB disk space for models and dependencies
94	CUDA-compatible GPU (optional, for accelerated processing)
95	ARM64 or x86_64 architecture
96
97
98	Implementation Strategy
99	The core processing pipeline requires thoughtful implementation of several key systems:
100
101	Frame extraction and analysis
102
103	Implement selective sampling based on visual change detection
104	Utilize region proposal networks for element identification
105
106
107	Speech processing
108
109	Apply time-domain speaker diarization
110	Implement context-aware transcription with domain adaptation
111
112
113	Visual element extraction
114
115	Develop whiteboard/diagram detection with boundary recognition
116	Implement reconstruction of visual elements into vector formats
117
118
119	Knowledge integration
120
121	Create hierarchical structure of extracted concepts
122	Generate relationship mappings between identified elements
123
124
125	Action item synthesis
126
127	Apply intent recognition for commitment identification
128	Implement priority scoring based on contextual importance
129
130
131
132	Each component should be implemented as a separate module with clear interfaces, allowing for independent testing and optimization.
133
134	Development Approach
135	When implementing PlanOpticon, consider these architectural principles:
136
137	Pipeline Architecture
138
139	Design processing stages that can operate independently
140	Implement data passing between stages using standardized formats
141	Enable parallelization where appropriate
142	Consider using Python's asyncio for I/O-bound operations
143
144
145	Performance Optimization
146
147	Implement batched processing for GPU acceleration
148	Use memory mapping for large video files
149	Consider JIT compilation for performance-critical sections
150	Profile and optimize bottlenecks systematically
151
152
153	Error Handling
154
155	Implement comprehensive exception handling
156	Design graceful degradation paths for each component
157	Provide detailed logging for troubleshooting
158	Consider retry mechanisms for transient failures
159
160
161	Testing Strategy
162
163	Create comprehensive unit tests for each module
164	Implement integration tests for end-to-end pipeline
165	Develop benchmark tests for performance evaluation
166	Use property-based testing for complex components
167
168
169
170	The implementation should maintain separation of concerns while ensuring efficient data flow between components. Consider using dependency injection patterns to improve testability and component isolation.
171
172	License
173	MIT License
174
175	Contact
176	For questions or contributions, please open an issue on GitHub or contact the maintainers at [email protected].
177
178	ELETED README_new.md

	--- README.md
	+++ README.md
	@@ -1,176 +1,101 @@
1	# PlanOpticon
2
3	AI-powered video analysis and knowledge extraction.
4
5	PlanOpticon processes video recordings into structured knowledge — transcripts, diagrams, action items, key points, and knowledge graphs. It auto-discovers available models across OpenAI, Anthropic, and Gemini, and produces rich multi-format output.
6
7	## Features
8
9	- Multi-provider AI — Auto-discovers and routes to the best available model across OpenAI, Anthropic, and Google Gemini
10	- Smart frame extraction — Change detection for transitions + periodic capture for slow-evolving content (document scrolling, screen shares)
11	- People frame filtering — OpenCV face detection automatically removes webcam/video conference frames, keeping only shared content
12	- Diagram extraction — Vision model classification detects flowcharts, architecture diagrams, charts, and whiteboards
13	- Knowledge graphs — Extracts entities and relationships, builds and merges knowledge graphs across videos
14	- Action item detection — Finds commitments, tasks, and follow-ups with assignees and deadlines
15	- Batch processing — Process entire folders of videos with merged knowledge graphs and cross-referencing
16	- Rich output — Markdown, HTML, PDF reports. Mermaid diagrams, SVG/PNG renderings, JSON manifests
17	- Cloud sources — Fetch videos from Google Drive and Dropbox shared folders
18	- Checkpoint/resume — Pipeline resumes from where it left off if interrupted
19	- Screengrab fallback — When extraction isn't perfect, captures frames with captions — something is always better than nothing
20
21	## Quick Start
22
23	```bash
24	# Install
25	pip install planopticon
26
27	# Analyze a single video
28	planopticon analyze -i meeting.mp4 -o ./output
29
30	# Process a folder of videos
31	planopticon batch -i ./recordings -o ./output --title "Weekly Meetings"
32
33	# See available AI models
34	planopticon list-models
35	```
36
37	## Installation
38
39	### From PyPI
40
41	```bash
42	pip install planopticon
43
44	# With all extras (PDF, cloud sources, GPU)
45	pip install planopticon[all]
46	```
47
48	### From Source
49
50	```bash
51	git clone https://github.com/ConflictHQ/PlanOpticon.git
52	cd PlanOpticon
53	pip install -e ".[dev]"
54	```
55
56	### Binary Download
57
58	Download standalone binaries (no Python required) from [GitHub Releases](https://github.com/ConflictHQ/PlanOpticon/releases).
59
60	### Requirements
61
62	- Python 3.10+
63	- FFmpeg (`brew install ffmpeg` / `apt install ffmpeg`)
64	- At least one API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY`
65
66	## Output Structure
67
68	```
69	output/
70	├── manifest.json # Single source of truth
71	├── transcript/
72	│ ├── transcript.json # Full transcript with timestamps
73	│ ├── transcript.txt # Plain text
74	│ └── transcript.srt # Subtitles
75	├── frames/ # Content frames (people filtered out)
76	├── diagrams/ # Detected diagrams + mermaid code
77	├── captures/ # Screengrab fallbacks
78	└── results/
79	├── analysis.md # Markdown report
80	├── analysis.html # HTML report
81	├── analysis.pdf # PDF report
82	├── knowledge_graph.json # Entities and relationships
83	├── key_points.json # Extracted key points
84	└── action_items.json # Tasks and follow-ups
85	```
86
87	## Processing Depth
88
89	\| Depth \| What you get \|
90	\|-------\|-------------\|
91	\| `basic` \| Transcription, key points, action items \|
92	\| `standard` \| + Diagram extraction (10 frames), knowledge graph, full reports \|
93	\| `comprehensive` \| + More frames analyzed (20), deeper extraction \|
94
95	## Documentation
96
97	Full documentation at [planopticon.dev](https://planopticon.dev)
98
99	## License
100
101	MIT License — Copyright (c) 2026 CONFLICT LLC











































































102
103	ELETED README_new.md

D README_new.md

-149

		--- a/README_new.md
		+++ b/README_new.md
		@@ -1,149 +0,0 @@
1		-# PlanOpticon
2		-
3		-Comprehensive Video Analysis & Knowledge Extraction CLI
4		-
5		-## Overview
6		-
7		-PlanOpticon is an advanced AI-powered CLI tool that conducts thorough analysis of video content, extracting structured knowledge, diagrams, and actionable insights. Using state-of-the-art computer vision and natural language processing techniques, PlanOpticon transforms video assets into valuable, structured information.
8		-
9		-## Core Features
10		-
11		-- Complete Transcription: Full speech-to-text with speaker attribution and semantic segmentation
12		-- Visual Element Extraction: Automated recognition and digitization of diagrams, charts, whiteboards, and visual aids
13		-- Action Item Detection: Intelligent identification and prioritization of tasks, commitments, and follow-ups
14		-- Knowledge Structure: Organization of extracted content into searchable, related concepts
15		-- Plan Generation: Synthesis of extracted elements into cohesive action plans and summaries
16		-
17		-## Installation
18		-
19		-### Prerequisites
20		-
21		-- Python 3.9+
22		-- FFmpeg (for audio/video processing)
23		-- API keys for cloud services (OpenAI, Google Cloud, etc.)
24		-
25		-### Setup
26		-
27		-1. Clone the repository:
28		-
29		-```bash
30		-git clone https://github.com/yourusername/planopticon.git
31		-cd planopticon
32		-```
33		-
34		-2. Run the setup script which creates a virtual environment and installs dependencies:
35		-
36		-```bash
37		-./scripts/setup.sh
38		-```
39		-
40		-3. Configure your API keys by editing the `.env` file created during setup.
41		-
42		-### Manual Installation
43		-
44		-If you prefer to set up manually:
45		-
46		-```bash
47		-# Create virtual environment
48		-python -m venv venv
49		-source venv/bin/activate # On Windows: venv\Scripts\activate
50		-
51		-# Install dependencies
52		-pip install -e .
53		-
54		-# Install optional GPU dependencies (if available)
55		-pip install -r requirements-gpu.txt # For NVIDIA GPUs
56		-pip install -r requirements-apple.txt # For Apple Silicon
57		-```
58		-
59		-## Usage
60		-
61		-PlanOpticon is designed as a command-line interface tool:
62		-
63		-```bash
64		-# Basic usage
65		-planopticon analyze --input video.mp4 --output analysis/
66		-
67		-# Specify processing depth
68		-planopticon analyze --input video.mp4 --depth comprehensive --output analysis/
69		-
70		-# Focus on specific extraction types
71		-planopticon analyze --input video.mp4 --focus "diagrams,action-items" --output analysis/
72		-
73		-# Process with GPU acceleration
74		-planopticon analyze --input video.mp4 --use-gpu --output analysis/
75		-```
76		-
77		-### Output Structure
78		-
79		-```
80		-analysis/
81		-├── transcript/
82		-│ ├── video_name.json # Full transcription with timestamps and speakers
83		-│ ├── video_name.txt # Plain text transcription
84		-│ └── video_name.srt # Subtitle format
85		-├── frames/ # Extracted key frames
86		-│ ├── frame_0001.jpg
87		-│ └── frame_0002.jpg
88		-├── audio/ # Extracted audio
89		-│ └── video_name.wav
90		-├── diagrams/ # Extracted and digitized visual elements
91		-│ ├── diagram_001.svg
92		-│ └── whiteboard_001.svg
93		-└── cache/ # API response cache
94		-```
95		-
96		-## Development
97		-
98		-### Architecture
99		-
100		-PlanOpticon follows a modular pipeline architecture:
101		-
102		-```
103		-video_processor/
104		-├── extractors/ # Video and audio extraction
105		-├── api/ # Cloud API integrations
106		-├── analyzers/ # Content analysis components
107		-├── integrators/ # Knowledge integration
108		-├── utils/ # Common utilities
109		-└── cli/ # Command-line interface
110		-```
111		-
112		-### Code Standards
113		-
114		-- Follow PEP 8 style guidelines for all Python code
115		-- Write comprehensive docstrings using NumPy style
116		-- Include type hints consistently throughout the codebase
117		-- Maintain test coverage for key components
118		-
119		-### Testing
120		-
121		-Run tests with pytest:
122		-
123		-```bash
124		-pytest
125		-```
126		-
127		-## System Requirements
128		-
129		-- Python 3.9+
130		-- 8GB RAM minimum (16GB recommended)
131		-- 2GB disk space for models and dependencies
132		-- CUDA-compatible GPU (optional, for accelerated processing)
133		-- ARM64 or x86_64 architecture
134		-
135		-## License
136		-
137		-MIT License
138		-
139		-## Roadmap
140		-
141		-See [work_plan.md](work_plan.md) for detailed development roadmap and milestones.
142		-
143		-## Contributing
144		-
145		-Contributions are welcome! Please feel free to submit a Pull Request.
146		-
147		-## Contact
148		-
149		-For questions or contributions, please open an issue on GitHub or contact the maintainers at [email protected].

	--- a/README_new.md
	+++ b/README_new.md
	@@ -1,149 +0,0 @@
1	# PlanOpticon
2
3	Comprehensive Video Analysis & Knowledge Extraction CLI
4
5	## Overview
6
7	PlanOpticon is an advanced AI-powered CLI tool that conducts thorough analysis of video content, extracting structured knowledge, diagrams, and actionable insights. Using state-of-the-art computer vision and natural language processing techniques, PlanOpticon transforms video assets into valuable, structured information.
8
9	## Core Features
10
11	- Complete Transcription: Full speech-to-text with speaker attribution and semantic segmentation
12	- Visual Element Extraction: Automated recognition and digitization of diagrams, charts, whiteboards, and visual aids
13	- Action Item Detection: Intelligent identification and prioritization of tasks, commitments, and follow-ups
14	- Knowledge Structure: Organization of extracted content into searchable, related concepts
15	- Plan Generation: Synthesis of extracted elements into cohesive action plans and summaries
16
17	## Installation
18
19	### Prerequisites
20
21	- Python 3.9+
22	- FFmpeg (for audio/video processing)
23	- API keys for cloud services (OpenAI, Google Cloud, etc.)
24
25	### Setup
26
27	1. Clone the repository:
28
29	```bash
30	git clone https://github.com/yourusername/planopticon.git
31	cd planopticon
32	```
33
34	2. Run the setup script which creates a virtual environment and installs dependencies:
35
36	```bash
37	./scripts/setup.sh
38	```
39
40	3. Configure your API keys by editing the `.env` file created during setup.
41
42	### Manual Installation
43
44	If you prefer to set up manually:
45
46	```bash
47	# Create virtual environment
48	python -m venv venv
49	source venv/bin/activate # On Windows: venv\Scripts\activate
50
51	# Install dependencies
52	pip install -e .
53
54	# Install optional GPU dependencies (if available)
55	pip install -r requirements-gpu.txt # For NVIDIA GPUs
56	pip install -r requirements-apple.txt # For Apple Silicon
57	```
58
59	## Usage
60
61	PlanOpticon is designed as a command-line interface tool:
62
63	```bash
64	# Basic usage
65	planopticon analyze --input video.mp4 --output analysis/
66
67	# Specify processing depth
68	planopticon analyze --input video.mp4 --depth comprehensive --output analysis/
69
70	# Focus on specific extraction types
71	planopticon analyze --input video.mp4 --focus "diagrams,action-items" --output analysis/
72
73	# Process with GPU acceleration
74	planopticon analyze --input video.mp4 --use-gpu --output analysis/
75	```
76
77	### Output Structure
78
79	```
80	analysis/
81	├── transcript/
82	│ ├── video_name.json # Full transcription with timestamps and speakers
83	│ ├── video_name.txt # Plain text transcription
84	│ └── video_name.srt # Subtitle format
85	├── frames/ # Extracted key frames
86	│ ├── frame_0001.jpg
87	│ └── frame_0002.jpg
88	├── audio/ # Extracted audio
89	│ └── video_name.wav
90	├── diagrams/ # Extracted and digitized visual elements
91	│ ├── diagram_001.svg
92	│ └── whiteboard_001.svg
93	└── cache/ # API response cache
94	```
95
96	## Development
97
98	### Architecture
99
100	PlanOpticon follows a modular pipeline architecture:
101
102	```
103	video_processor/
104	├── extractors/ # Video and audio extraction
105	├── api/ # Cloud API integrations
106	├── analyzers/ # Content analysis components
107	├── integrators/ # Knowledge integration
108	├── utils/ # Common utilities
109	└── cli/ # Command-line interface
110	```
111
112	### Code Standards
113
114	- Follow PEP 8 style guidelines for all Python code
115	- Write comprehensive docstrings using NumPy style
116	- Include type hints consistently throughout the codebase
117	- Maintain test coverage for key components
118
119	### Testing
120
121	Run tests with pytest:
122
123	```bash
124	pytest
125	```
126
127	## System Requirements
128
129	- Python 3.9+
130	- 8GB RAM minimum (16GB recommended)
131	- 2GB disk space for models and dependencies
132	- CUDA-compatible GPU (optional, for accelerated processing)
133	- ARM64 or x86_64 architecture
134
135	## License
136
137	MIT License
138
139	## Roadmap
140
141	See [work_plan.md](work_plan.md) for detailed development roadmap and milestones.
142
143	## Contributing
144
145	Contributions are welcome! Please feel free to submit a Pull Request.
146
147	## Contact
148
149	For questions or contributions, please open an issue on GitHub or contact the maintainers at [email protected].

	--- a/README_new.md
	+++ b/README_new.md
	@@ -1,149 +0,0 @@

M docs/cli-reference.md

+50

		--- docs/cli-reference.md
		+++ docs/cli-reference.md
		@@ -41,10 +41,14 @@
41	41	\| `--pattern` \| TEXT \| `.mp4,.mkv,.avi,.mov,*.webm` \| File glob patterns \|
42	42	\| `--title` \| TEXT \| `Batch Processing Results` \| Batch title \|
43	43	\| `-p`, `--provider` \| `auto\\|openai\\|anthropic\\|gemini` \| `auto` \| API provider \|
44	44	\| `--vision-model` \| TEXT \| auto \| Override vision model \|
45	45	\| `--chat-model` \| TEXT \| auto \| Override chat model \|
	46	+\| `--source` \| `local\\|gdrive\\|dropbox` \| `local` \| Video source \|
	47	+\| `--folder-id` \| TEXT \| none \| Google Drive folder ID \|
	48	+\| `--folder-path` \| TEXT \| none \| Cloud folder path \|
	49	+\| `--recursive/--no-recursive` \| FLAG \| recursive \| Recurse into subfolders \|
46	50
47	51	---
48	52
49	53	## `planopticon list-models`
50	54
		@@ -70,14 +74,60 @@
70	74	\|--------\|------\|---------\|-------------\|
71	75	\| `--cache-dir` \| PATH \| `$CACHE_DIR` \| Path to cache directory \|
72	76	\| `--older-than` \| INT \| all \| Clear entries older than N seconds \|
73	77	\| `--all` \| FLAG \| off \| Clear all cache entries \|
74	78
	79	+---
	80	+
	81	+## `planopticon agent-analyze`
	82	+
	83	+Agentic video analysis — adaptive, intelligent processing that adjusts depth and focus based on content.
	84	+
	85	+```bash
	86	+planopticon agent-analyze [OPTIONS]
	87	+```
	88	+
	89	+\| Option \| Type \| Default \| Description \|
	90	+\|--------\|------\|---------\|-------------\|
	91	+\| `-i`, `--input` \| PATH \| required \| Input video file path \|
	92	+\| `-o`, `--output` \| PATH \| required \| Output directory \|
	93	+\| `--depth` \| `basic\\|standard\\|comprehensive` \| `standard` \| Initial processing depth (agent may adapt) \|
	94	+\| `--title` \| TEXT \| auto \| Report title \|
	95	+\| `-p`, `--provider` \| `auto\\|openai\\|anthropic\\|gemini` \| `auto` \| API provider \|
	96	+\| `--vision-model` \| TEXT \| auto \| Override vision model \|
	97	+\| `--chat-model` \| TEXT \| auto \| Override chat model \|
	98	+
	99	+---
	100	+
	101	+## `planopticon auth`
	102	+
	103	+Authenticate with a cloud storage service for batch processing.
	104	+
	105	+```bash
	106	+planopticon auth SERVICE
	107	+```
	108	+
	109	+\| Argument \| Values \| Description \|
	110	+\|----------\|--------\|-------------\|
	111	+\| `SERVICE` \| `google\\|dropbox` \| Cloud service to authenticate with \|
	112	+
	113	+Examples:
	114	+
	115	+```bash
	116	+# Authenticate with Google Drive (interactive OAuth2)
	117	+planopticon auth google
	118	+
	119	+# Authenticate with Dropbox
	120	+planopticon auth dropbox
	121	+```
	122	+
	123	+After authentication, use `planopticon batch --source gdrive` or `--source dropbox` to process cloud videos.
	124	+
75	125	---
76	126
77	127	## Global options
78	128
79	129	\| Option \| Description \|
80	130	\|--------\|-------------\|
81	131	\| `-v`, `--verbose` \| Enable debug-level logging \|
82	132	\| `--version` \| Show version and exit \|
83	133	\| `--help` \| Show help and exit \|
84	134

	--- docs/cli-reference.md
	+++ docs/cli-reference.md
	@@ -41,10 +41,14 @@
41	\| `--pattern` \| TEXT \| `.mp4,.mkv,.avi,.mov,*.webm` \| File glob patterns \|
42	\| `--title` \| TEXT \| `Batch Processing Results` \| Batch title \|
43	\| `-p`, `--provider` \| `auto\\|openai\\|anthropic\\|gemini` \| `auto` \| API provider \|
44	\| `--vision-model` \| TEXT \| auto \| Override vision model \|
45	\| `--chat-model` \| TEXT \| auto \| Override chat model \|




46
47	---
48
49	## `planopticon list-models`
50
	@@ -70,14 +74,60 @@
70	\|--------\|------\|---------\|-------------\|
71	\| `--cache-dir` \| PATH \| `$CACHE_DIR` \| Path to cache directory \|
72	\| `--older-than` \| INT \| all \| Clear entries older than N seconds \|
73	\| `--all` \| FLAG \| off \| Clear all cache entries \|
74














































75	---
76
77	## Global options
78
79	\| Option \| Description \|
80	\|--------\|-------------\|
81	\| `-v`, `--verbose` \| Enable debug-level logging \|
82	\| `--version` \| Show version and exit \|
83	\| `--help` \| Show help and exit \|
84

	--- docs/cli-reference.md
	+++ docs/cli-reference.md
	@@ -41,10 +41,14 @@
41	\| `--pattern` \| TEXT \| `.mp4,.mkv,.avi,.mov,*.webm` \| File glob patterns \|
42	\| `--title` \| TEXT \| `Batch Processing Results` \| Batch title \|
43	\| `-p`, `--provider` \| `auto\\|openai\\|anthropic\\|gemini` \| `auto` \| API provider \|
44	\| `--vision-model` \| TEXT \| auto \| Override vision model \|
45	\| `--chat-model` \| TEXT \| auto \| Override chat model \|
46	\| `--source` \| `local\\|gdrive\\|dropbox` \| `local` \| Video source \|
47	\| `--folder-id` \| TEXT \| none \| Google Drive folder ID \|
48	\| `--folder-path` \| TEXT \| none \| Cloud folder path \|
49	\| `--recursive/--no-recursive` \| FLAG \| recursive \| Recurse into subfolders \|
50
51	---
52
53	## `planopticon list-models`
54
	@@ -70,14 +74,60 @@
74	\|--------\|------\|---------\|-------------\|
75	\| `--cache-dir` \| PATH \| `$CACHE_DIR` \| Path to cache directory \|
76	\| `--older-than` \| INT \| all \| Clear entries older than N seconds \|
77	\| `--all` \| FLAG \| off \| Clear all cache entries \|
78
79	---
80
81	## `planopticon agent-analyze`
82
83	Agentic video analysis — adaptive, intelligent processing that adjusts depth and focus based on content.
84
85	```bash
86	planopticon agent-analyze [OPTIONS]
87	```
88
89	\| Option \| Type \| Default \| Description \|
90	\|--------\|------\|---------\|-------------\|
91	\| `-i`, `--input` \| PATH \| required \| Input video file path \|
92	\| `-o`, `--output` \| PATH \| required \| Output directory \|
93	\| `--depth` \| `basic\\|standard\\|comprehensive` \| `standard` \| Initial processing depth (agent may adapt) \|
94	\| `--title` \| TEXT \| auto \| Report title \|
95	\| `-p`, `--provider` \| `auto\\|openai\\|anthropic\\|gemini` \| `auto` \| API provider \|
96	\| `--vision-model` \| TEXT \| auto \| Override vision model \|
97	\| `--chat-model` \| TEXT \| auto \| Override chat model \|
98
99	---
100
101	## `planopticon auth`
102
103	Authenticate with a cloud storage service for batch processing.
104
105	```bash
106	planopticon auth SERVICE
107	```
108
109	\| Argument \| Values \| Description \|
110	\|----------\|--------\|-------------\|
111	\| `SERVICE` \| `google\\|dropbox` \| Cloud service to authenticate with \|
112
113	Examples:
114
115	```bash
116	# Authenticate with Google Drive (interactive OAuth2)
117	planopticon auth google
118
119	# Authenticate with Dropbox
120	planopticon auth dropbox
121	```
122
123	After authentication, use `planopticon batch --source gdrive` or `--source dropbox` to process cloud videos.
124
125	---
126
127	## Global options
128
129	\| Option \| Description \|
130	\|--------\|-------------\|
131	\| `-v`, `--verbose` \| Enable debug-level logging \|
132	\| `--version` \| Show version and exit \|
133	\| `--help` \| Show help and exit \|
134

M docs/getting-started/configuration.md

+10 -1

		--- docs/getting-started/configuration.md
		+++ docs/getting-started/configuration.md
		@@ -30,13 +30,22 @@
30	30	# Sample rate: frames per second (default: 0.5)
31	31	planopticon analyze -i video.mp4 -o ./out --sampling-rate 1.0
32	32
33	33	# Change threshold: visual difference needed to keep a frame (default: 0.15)
34	34	planopticon analyze -i video.mp4 -o ./out --change-threshold 0.1
	35	+
	36	+# Periodic capture: capture a frame every N seconds regardless of change (default: 30)
	37	+# Useful for slow-evolving content like document scrolling
	38	+planopticon analyze -i video.mp4 -o ./out --periodic-capture 15
	39	+
	40	+# Disable periodic capture (rely only on change detection)
	41	+planopticon analyze -i video.mp4 -o ./out --periodic-capture 0
35	42	```
36	43
37		-Lower `change-threshold` = more frames kept. Higher `sampling-rate` = more candidates.
	44	+Lower `change-threshold` = more frames kept. Higher `sampling-rate` = more candidates. Periodic capture catches content that changes too slowly for change detection (e.g., scrolling through a document during a screen share).
	45	+
	46	+People/webcam frames are automatically filtered out using face detection — no configuration needed.
38	47
39	48	## Focus areas
40	49
41	50	Limit processing to specific extraction types:
42	51
43	52

	--- docs/getting-started/configuration.md
	+++ docs/getting-started/configuration.md
	@@ -30,13 +30,22 @@
30	# Sample rate: frames per second (default: 0.5)
31	planopticon analyze -i video.mp4 -o ./out --sampling-rate 1.0
32
33	# Change threshold: visual difference needed to keep a frame (default: 0.15)
34	planopticon analyze -i video.mp4 -o ./out --change-threshold 0.1







35	```
36
37	Lower `change-threshold` = more frames kept. Higher `sampling-rate` = more candidates.


38
39	## Focus areas
40
41	Limit processing to specific extraction types:
42
43

	--- docs/getting-started/configuration.md
	+++ docs/getting-started/configuration.md
	@@ -30,13 +30,22 @@
30	# Sample rate: frames per second (default: 0.5)
31	planopticon analyze -i video.mp4 -o ./out --sampling-rate 1.0
32
33	# Change threshold: visual difference needed to keep a frame (default: 0.15)
34	planopticon analyze -i video.mp4 -o ./out --change-threshold 0.1
35
36	# Periodic capture: capture a frame every N seconds regardless of change (default: 30)
37	# Useful for slow-evolving content like document scrolling
38	planopticon analyze -i video.mp4 -o ./out --periodic-capture 15
39
40	# Disable periodic capture (rely only on change detection)
41	planopticon analyze -i video.mp4 -o ./out --periodic-capture 0
42	```
43
44	Lower `change-threshold` = more frames kept. Higher `sampling-rate` = more candidates. Periodic capture catches content that changes too slowly for change detection (e.g., scrolling through a document during a screen share).
45
46	People/webcam frames are automatically filtered out using face detection — no configuration needed.
47
48	## Focus areas
49
50	Limit processing to specific extraction types:
51
52

M docs/guide/cloud-sources.md

-3

		--- docs/guide/cloud-sources.md
		+++ docs/guide/cloud-sources.md
		@@ -1,10 +1,7 @@
1	1	# Cloud Sources
2	2
3		-!!! note "Coming soon"
4		- Cloud source integration is under active development.
5		-
6	3	PlanOpticon can fetch videos directly from cloud storage services.
7	4
8	5	## Google Drive
9	6
10	7	### Service account auth
11	8

	--- docs/guide/cloud-sources.md
	+++ docs/guide/cloud-sources.md
	@@ -1,10 +1,7 @@
1	# Cloud Sources
2
3	!!! note "Coming soon"
4	Cloud source integration is under active development.
5
6	PlanOpticon can fetch videos directly from cloud storage services.
7
8	## Google Drive
9
10	### Service account auth
11

	--- docs/guide/cloud-sources.md
	+++ docs/guide/cloud-sources.md
	@@ -1,10 +1,7 @@
1	# Cloud Sources
2



3	PlanOpticon can fetch videos directly from cloud storage services.
4
5	## Google Drive
6
7	### Service account auth
8

M docs/index.md

+5 -2

		--- docs/index.md
		+++ docs/index.md
		@@ -7,16 +7,19 @@
7	7	---
8	8
9	9	## Features
10	10
11	11	- Multi-provider AI — Automatically discovers and routes to the best available model across OpenAI, Anthropic, and Google Gemini
	12	+- Smart frame extraction — Change detection for transitions + periodic capture (every 30s) for slow-evolving content like document scrolling
	13	+- People frame filtering — OpenCV face detection removes webcam/video conference frames, keeping only shared content (slides, documents, screen shares)
12	14	- Diagram extraction — Vision model-based classification detects flowcharts, architecture diagrams, charts, and whiteboards
13	15	- Knowledge graphs — Extracts entities and relationships, builds and merges knowledge graphs across videos
14	16	- Action item detection — Finds commitments, tasks, and follow-ups with assignees and deadlines
15	17	- Batch processing — Process entire folders of videos with merged knowledge graphs and cross-referencing
16	18	- Rich output — Markdown, HTML, PDF, Mermaid diagrams, SVG/PNG renderings, JSON manifests
17	19	- Cloud sources — Fetch videos from Google Drive and Dropbox shared folders
	20	+- Checkpoint/resume — Pipeline resumes from where it left off if interrupted — no wasted work
18	21	- Screengrab fallback — When extraction isn't perfect, captures frames with captions — something is always better than nothing
19	22
20	23	## Quick Start
21	24
22	25	```bash
		@@ -54,16 +57,16 @@
54	57	```
55	58
56	59	=== "Binary (no Python needed)"
57	60
58	61	Download the latest binary for your platform from
59		- [GitHub Releases](https://github.com/conflict-llc/PlanOpticon/releases).
	62	+ [GitHub Releases](https://github.com/ConflictHQ/PlanOpticon/releases).
60	63
61	64	## Requirements
62	65
63	66	- Python 3.10+
64	67	- FFmpeg (for audio extraction)
65	68	- At least one API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY`
66	69
67	70	## License
68	71
69		-MIT License — Copyright (c) 2025 CONFLICT LLC. All rights reserved.
	72	+MIT License — Copyright (c) 2026 CONFLICT LLC. All rights reserved.
70	73

	--- docs/index.md
	+++ docs/index.md
	@@ -7,16 +7,19 @@
7	---
8
9	## Features
10
11	- Multi-provider AI — Automatically discovers and routes to the best available model across OpenAI, Anthropic, and Google Gemini


12	- Diagram extraction — Vision model-based classification detects flowcharts, architecture diagrams, charts, and whiteboards
13	- Knowledge graphs — Extracts entities and relationships, builds and merges knowledge graphs across videos
14	- Action item detection — Finds commitments, tasks, and follow-ups with assignees and deadlines
15	- Batch processing — Process entire folders of videos with merged knowledge graphs and cross-referencing
16	- Rich output — Markdown, HTML, PDF, Mermaid diagrams, SVG/PNG renderings, JSON manifests
17	- Cloud sources — Fetch videos from Google Drive and Dropbox shared folders

18	- Screengrab fallback — When extraction isn't perfect, captures frames with captions — something is always better than nothing
19
20	## Quick Start
21
22	```bash
	@@ -54,16 +57,16 @@
54	```
55
56	=== "Binary (no Python needed)"
57
58	Download the latest binary for your platform from
59	[GitHub Releases](https://github.com/conflict-llc/PlanOpticon/releases).
60
61	## Requirements
62
63	- Python 3.10+
64	- FFmpeg (for audio extraction)
65	- At least one API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY`
66
67	## License
68
69	MIT License — Copyright (c) 2025 CONFLICT LLC. All rights reserved.
70

	--- docs/index.md
	+++ docs/index.md
	@@ -7,16 +7,19 @@
7	---
8
9	## Features
10
11	- Multi-provider AI — Automatically discovers and routes to the best available model across OpenAI, Anthropic, and Google Gemini
12	- Smart frame extraction — Change detection for transitions + periodic capture (every 30s) for slow-evolving content like document scrolling
13	- People frame filtering — OpenCV face detection removes webcam/video conference frames, keeping only shared content (slides, documents, screen shares)
14	- Diagram extraction — Vision model-based classification detects flowcharts, architecture diagrams, charts, and whiteboards
15	- Knowledge graphs — Extracts entities and relationships, builds and merges knowledge graphs across videos
16	- Action item detection — Finds commitments, tasks, and follow-ups with assignees and deadlines
17	- Batch processing — Process entire folders of videos with merged knowledge graphs and cross-referencing
18	- Rich output — Markdown, HTML, PDF, Mermaid diagrams, SVG/PNG renderings, JSON manifests
19	- Cloud sources — Fetch videos from Google Drive and Dropbox shared folders
20	- Checkpoint/resume — Pipeline resumes from where it left off if interrupted — no wasted work
21	- Screengrab fallback — When extraction isn't perfect, captures frames with captions — something is always better than nothing
22
23	## Quick Start
24
25	```bash
	@@ -54,16 +57,16 @@
57	```
58
59	=== "Binary (no Python needed)"
60
61	Download the latest binary for your platform from
62	[GitHub Releases](https://github.com/ConflictHQ/PlanOpticon/releases).
63
64	## Requirements
65
66	- Python 3.10+
67	- FFmpeg (for audio extraction)
68	- At least one API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY`
69
70	## License
71
72	MIT License — Copyright (c) 2026 CONFLICT LLC. All rights reserved.
73

M mkdocs.yml

+1 -1

		--- mkdocs.yml
		+++ mkdocs.yml
		@@ -88,6 +88,6 @@
88	88	extra:
89	89	social:
90	90	- icon: fontawesome/brands/github
91	91	link: https://github.com/conflict-llc/PlanOpticon
92	92
93		-copyright: Copyright © 2025 CONFLICT LLC
	93	+copyright: Copyright © 2026 CONFLICT LLC
94	94

	--- mkdocs.yml
	+++ mkdocs.yml
	@@ -88,6 +88,6 @@
88	extra:
89	social:
90	- icon: fontawesome/brands/github
91	link: https://github.com/conflict-llc/PlanOpticon
92
93	copyright: Copyright © 2025 CONFLICT LLC
94

	--- mkdocs.yml
	+++ mkdocs.yml
	@@ -88,6 +88,6 @@
88	extra:
89	social:
90	- icon: fontawesome/brands/github
91	link: https://github.com/conflict-llc/PlanOpticon
92
93	copyright: Copyright © 2026 CONFLICT LLC
94

M video_processor/cli/commands.py

+103 -1

		--- video_processor/cli/commands.py
		+++ video_processor/cli/commands.py
		@@ -34,19 +34,22 @@
34	34	for handler in root_logger.handlers:
35	35	root_logger.removeHandler(handler)
36	36	root_logger.addHandler(console_handler)
37	37
38	38
39		-@click.group()
	39	+@click.group(invoke_without_command=True)
40	40	@click.option("--verbose", "-v", is_flag=True, help="Enable verbose output")
41	41	@click.version_option("0.2.0", prog_name="PlanOpticon")
42	42	@click.pass_context
43	43	def cli(ctx, verbose):
44	44	"""PlanOpticon - Comprehensive Video Analysis & Knowledge Extraction Tool."""
45	45	ctx.ensure_object(dict)
46	46	ctx.obj["verbose"] = verbose
47	47	setup_logging(verbose)
	48	+
	49	+ if ctx.invoked_subcommand is None:
	50	+ _interactive_menu(ctx)
48	51
49	52
50	53	@cli.command()
51	54	@click.option("--input", "-i", required=True, type=click.Path(exists=True), help="Input video file path")
52	55	@click.option("--output", "-o", required=True, type=click.Path(), help="Output directory")
		@@ -450,13 +453,112 @@
450	453	click.echo("Dropbox authentication successful.")
451	454	else:
452	455	click.echo("Dropbox authentication failed.", err=True)
453	456	sys.exit(1)
454	457
	458	+
	459	+def _interactive_menu(ctx):
	460	+ """Show an interactive menu when planopticon is run with no arguments."""
	461	+ click.echo()
	462	+ click.echo(" PlanOpticon v0.2.0")
	463	+ click.echo(" Comprehensive Video Analysis & Knowledge Extraction")
	464	+ click.echo()
	465	+ click.echo(" 1. Analyze a video")
	466	+ click.echo(" 2. Batch process a folder")
	467	+ click.echo(" 3. List available models")
	468	+ click.echo(" 4. Authenticate cloud service")
	469	+ click.echo(" 5. Clear cache")
	470	+ click.echo(" 6. Show help")
	471	+ click.echo()
	472	+
	473	+ choice = click.prompt(" Select an option", type=click.IntRange(1, 6))
	474	+
	475	+ if choice == 1:
	476	+ input_path = click.prompt(" Video file path", type=click.Path(exists=True))
	477	+ output_dir = click.prompt(" Output directory", type=click.Path())
	478	+ depth = click.prompt(
	479	+ " Processing depth",
	480	+ type=click.Choice(["basic", "standard", "comprehensive"]),
	481	+ default="standard",
	482	+ )
	483	+ provider = click.prompt(
	484	+ " Provider",
	485	+ type=click.Choice(["auto", "openai", "anthropic", "gemini"]),
	486	+ default="auto",
	487	+ )
	488	+ ctx.invoke(
	489	+ analyze,
	490	+ input=input_path,
	491	+ output=output_dir,
	492	+ depth=depth,
	493	+ focus=None,
	494	+ use_gpu=False,
	495	+ sampling_rate=0.5,
	496	+ change_threshold=0.15,
	497	+ periodic_capture=30.0,
	498	+ title=None,
	499	+ provider=provider,
	500	+ vision_model=None,
	501	+ chat_model=None,
	502	+ )
	503	+
	504	+ elif choice == 2:
	505	+ input_dir = click.prompt(" Video directory", type=click.Path(exists=True))
	506	+ output_dir = click.prompt(" Output directory", type=click.Path())
	507	+ depth = click.prompt(
	508	+ " Processing depth",
	509	+ type=click.Choice(["basic", "standard", "comprehensive"]),
	510	+ default="standard",
	511	+ )
	512	+ provider = click.prompt(
	513	+ " Provider",
	514	+ type=click.Choice(["auto", "openai", "anthropic", "gemini"]),
	515	+ default="auto",
	516	+ )
	517	+ ctx.invoke(
	518	+ batch,
	519	+ input_dir=input_dir,
	520	+ output=output_dir,
	521	+ depth=depth,
	522	+ pattern=".mp4,.mkv,.avi,.mov,*.webm",
	523	+ title="Batch Processing Results",
	524	+ provider=provider,
	525	+ vision_model=None,
	526	+ chat_model=None,
	527	+ source="local",
	528	+ folder_id=None,
	529	+ folder_path=None,
	530	+ recursive=True,
	531	+ )
	532	+
	533	+ elif choice == 3:
	534	+ ctx.invoke(list_models)
	535	+
	536	+ elif choice == 4:
	537	+ service = click.prompt(
	538	+ " Cloud service",
	539	+ type=click.Choice(["google", "dropbox"]),
	540	+ )
	541	+ ctx.invoke(auth, service=service)
	542	+
	543	+ elif choice == 5:
	544	+ cache_dir = click.prompt(" Cache directory path", type=click.Path())
	545	+ clear_all = click.confirm(" Clear all entries?", default=True)
	546	+ ctx.invoke(
	547	+ clear_cache,
	548	+ cache_dir=cache_dir,
	549	+ older_than=None,
	550	+ clear_all=clear_all,
	551	+ )
	552	+
	553	+ elif choice == 6:
	554	+ click.echo()
	555	+ click.echo(ctx.get_help())
	556	+
455	557
456	558	def main():
457	559	"""Entry point for command-line usage."""
458	560	cli(obj={})
459	561
460	562
461	563	if __name__ == "__main__":
462	564	main()
463	565

	--- video_processor/cli/commands.py
	+++ video_processor/cli/commands.py
	@@ -34,19 +34,22 @@
34	for handler in root_logger.handlers:
35	root_logger.removeHandler(handler)
36	root_logger.addHandler(console_handler)
37
38
39	@click.group()
40	@click.option("--verbose", "-v", is_flag=True, help="Enable verbose output")
41	@click.version_option("0.2.0", prog_name="PlanOpticon")
42	@click.pass_context
43	def cli(ctx, verbose):
44	"""PlanOpticon - Comprehensive Video Analysis & Knowledge Extraction Tool."""
45	ctx.ensure_object(dict)
46	ctx.obj["verbose"] = verbose
47	setup_logging(verbose)



48
49
50	@cli.command()
51	@click.option("--input", "-i", required=True, type=click.Path(exists=True), help="Input video file path")
52	@click.option("--output", "-o", required=True, type=click.Path(), help="Output directory")
	@@ -450,13 +453,112 @@
450	click.echo("Dropbox authentication successful.")
451	else:
452	click.echo("Dropbox authentication failed.", err=True)
453	sys.exit(1)
454



































































































455
456	def main():
457	"""Entry point for command-line usage."""
458	cli(obj={})
459
460
461	if __name__ == "__main__":
462	main()
463

	--- video_processor/cli/commands.py
	+++ video_processor/cli/commands.py
	@@ -34,19 +34,22 @@
34	for handler in root_logger.handlers:
35	root_logger.removeHandler(handler)
36	root_logger.addHandler(console_handler)
37
38
39	@click.group(invoke_without_command=True)
40	@click.option("--verbose", "-v", is_flag=True, help="Enable verbose output")
41	@click.version_option("0.2.0", prog_name="PlanOpticon")
42	@click.pass_context
43	def cli(ctx, verbose):
44	"""PlanOpticon - Comprehensive Video Analysis & Knowledge Extraction Tool."""
45	ctx.ensure_object(dict)
46	ctx.obj["verbose"] = verbose
47	setup_logging(verbose)
48
49	if ctx.invoked_subcommand is None:
50	_interactive_menu(ctx)
51
52
53	@cli.command()
54	@click.option("--input", "-i", required=True, type=click.Path(exists=True), help="Input video file path")
55	@click.option("--output", "-o", required=True, type=click.Path(), help="Output directory")
	@@ -450,13 +453,112 @@
453	click.echo("Dropbox authentication successful.")
454	else:
455	click.echo("Dropbox authentication failed.", err=True)
456	sys.exit(1)
457
458
459	def _interactive_menu(ctx):
460	"""Show an interactive menu when planopticon is run with no arguments."""
461	click.echo()
462	click.echo(" PlanOpticon v0.2.0")
463	click.echo(" Comprehensive Video Analysis & Knowledge Extraction")
464	click.echo()
465	click.echo(" 1. Analyze a video")
466	click.echo(" 2. Batch process a folder")
467	click.echo(" 3. List available models")
468	click.echo(" 4. Authenticate cloud service")
469	click.echo(" 5. Clear cache")
470	click.echo(" 6. Show help")
471	click.echo()
472
473	choice = click.prompt(" Select an option", type=click.IntRange(1, 6))
474
475	if choice == 1:
476	input_path = click.prompt(" Video file path", type=click.Path(exists=True))
477	output_dir = click.prompt(" Output directory", type=click.Path())
478	depth = click.prompt(
479	" Processing depth",
480	type=click.Choice(["basic", "standard", "comprehensive"]),
481	default="standard",
482	)
483	provider = click.prompt(
484	" Provider",
485	type=click.Choice(["auto", "openai", "anthropic", "gemini"]),
486	default="auto",
487	)
488	ctx.invoke(
489	analyze,
490	input=input_path,
491	output=output_dir,
492	depth=depth,
493	focus=None,
494	use_gpu=False,
495	sampling_rate=0.5,
496	change_threshold=0.15,
497	periodic_capture=30.0,
498	title=None,
499	provider=provider,
500	vision_model=None,
501	chat_model=None,
502	)
503
504	elif choice == 2:
505	input_dir = click.prompt(" Video directory", type=click.Path(exists=True))
506	output_dir = click.prompt(" Output directory", type=click.Path())
507	depth = click.prompt(
508	" Processing depth",
509	type=click.Choice(["basic", "standard", "comprehensive"]),
510	default="standard",
511	)
512	provider = click.prompt(
513	" Provider",
514	type=click.Choice(["auto", "openai", "anthropic", "gemini"]),
515	default="auto",
516	)
517	ctx.invoke(
518	batch,
519	input_dir=input_dir,
520	output=output_dir,
521	depth=depth,
522	pattern=".mp4,.mkv,.avi,.mov,*.webm",
523	title="Batch Processing Results",
524	provider=provider,
525	vision_model=None,
526	chat_model=None,
527	source="local",
528	folder_id=None,
529	folder_path=None,
530	recursive=True,
531	)
532
533	elif choice == 3:
534	ctx.invoke(list_models)
535
536	elif choice == 4:
537	service = click.prompt(
538	" Cloud service",
539	type=click.Choice(["google", "dropbox"]),
540	)
541	ctx.invoke(auth, service=service)
542
543	elif choice == 5:
544	cache_dir = click.prompt(" Cache directory path", type=click.Path())
545	clear_all = click.confirm(" Clear all entries?", default=True)
546	ctx.invoke(
547	clear_cache,
548	cache_dir=cache_dir,
549	older_than=None,
550	clear_all=clear_all,
551	)
552
553	elif choice == 6:
554	click.echo()
555	click.echo(ctx.get_help())
556
557
558	def main():
559	"""Entry point for command-line usage."""
560	cli(obj={})
561
562
563	if __name__ == "__main__":
564	main()
565

PlanOpticon

Keyboard Shortcuts