PlanOpticon
Quick Start - PlanOpticon
Back to top
PlanOpticon
Quick Start
Quick Start¶
Analyze a single video¶
This runs the full pipeline:
- Extracts video frames (smart sampling, change detection)
- Extracts and transcribes audio
- Detects and analyzes diagrams, charts, whiteboards
- Builds a knowledge graph of entities and relationships
- Extracts key points and action items
- Generates markdown, HTML, and PDF reports
- Outputs a
manifest.jsonwith everything
Processing depth¶
# Quick scan — transcription + key points only
planopticon analyze -i video.mp4 -o ./out --depth basic
# Standard — includes diagram extraction (default)
planopticon analyze -i video.mp4 -o ./out --depth standard
# Deep — more frames analyzed, richer extraction
planopticon analyze -i video.mp4 -o ./out --depth comprehensive
Choose a provider¶
# Auto-detect best available (default)
planopticon analyze -i video.mp4 -o ./out
# Force a specific provider
planopticon analyze -i video.mp4 -o ./out --provider openai
# Use Ollama for fully offline processing (no API keys needed)
planopticon analyze -i video.mp4 -o ./out --provider ollama
# Override specific models
planopticon analyze -i video.mp4 -o ./out \
--vision-model gpt-4o \
--chat-model claude-sonnet-4-5-20250929
Batch processing¶
# Process all videos in a folder
planopticon batch -i ./recordings -o ./output
# Custom file patterns
planopticon batch -i ./recordings -o ./output --pattern "*.mp4,*.mov"
# With a title for the batch report
planopticon batch -i ./recordings -o ./output --title "Q4 Sprint Reviews"
Batch mode produces per-video outputs plus:
- Merged knowledge graph across all videos
- Batch summary with aggregated action items
- Cross-referenced entities
Ingest documents¶
Build a knowledge graph from documents, notes, or any text content:
# Ingest a single file
planopticon ingest ./meeting-notes.md --output ./kb
# Ingest a directory recursively
planopticon ingest ./docs/ --output ./kb --recursive
# Ingest from a URL
planopticon ingest "https://www.youtube.com/watch?v=example" --output ./kb
Companion REPL¶
Chat with your knowledge base interactively:
# Start the companion
planopticon companion --kb ./kb
# Use a specific provider
planopticon companion --kb ./kb --provider anthropic
The companion understands your knowledge graph and can answer questions, find connections, and summarize topics conversationally.
Planning agent¶
Run the planning agent for adaptive, goal-directed analysis:
# Interactive mode — the agent asks before each action
planopticon agent --kb ./kb --interactive
# Non-interactive with export
planopticon agent --kb ./kb --export ./plan.md
Query the knowledge graph¶
Query your knowledge graph directly without an AI provider:
# Show graph stats (entity/relationship counts)
planopticon query stats
# List entities by type
planopticon query "entities --type technology"
# Find neighbors of an entity
planopticon query "neighbors Alice"
# Natural language query (requires API key)
planopticon query "What technologies were discussed?"
# Interactive REPL
planopticon query -I
Export¶
Export your knowledge base to various formats:
# Export to Markdown files
planopticon export markdown --input ./kb --output ./docs
# Export to an Obsidian vault
planopticon export obsidian --input ./kb --output ~/Obsidian/PlanOpticon
# Export to Notion
planopticon export notion --input ./kb --parent-page abc123
# Export as exchange format (portable JSON)
planopticon export exchange --input ./kb --output ./export.json
Discover available models¶
Shows all models from configured providers with their capabilities (vision, chat, transcription).
Output structure¶
After processing, your output directory looks like:
output/
├── manifest.json # Single source of truth
├── transcript/
│ ├── transcript.json # Full transcript with segments
│ ├── transcript.txt # Plain text
│ └── transcript.srt # Subtitles
├── frames/ # Extracted video frames
├── diagrams/ # Detected diagrams + mermaid/SVG/PNG
├── captures/ # Screengrab fallbacks
└── results/
├── analysis.md # Markdown report
├── analysis.html # HTML report
├── analysis.pdf # PDF report
├── knowledge_graph.json
├── key_points.json
└── action_items.json