PlanOpticon

planopticon / docs / index.md

Source Rendered

Blame History Raw 93 lines

1	`# PlanOpticon`
2
3	`AI-powered video analysis, knowledge extraction, and planning.`
4
5	`PlanOpticon processes video recordings and documents into structured knowledge — transcripts, diagrams, action items, key points, and knowledge graphs. It connects to 20+ source platforms, auto-discovers available models across multiple AI providers, and produces rich multi-format output with an interactive companion REPL and planning agent.`
6
7	`---`
8
9	`## Features`
10
11	`- Multi-provider AI — Automatically discovers and routes to the best available model across OpenAI, Anthropic, Google Gemini, and more`
12	`- Planning agent — Agentic analysis that adaptively adjusts depth, focus, and strategy based on content`
13	`- Companion REPL — Interactive chat interface for exploring your knowledge base conversationally`
14	`- 20+ source connectors — Google Workspace, Microsoft 365, Zoom, Teams, Meet, Notion, GitHub, YouTube, Obsidian, Apple Notes, and more`
15	`- Document export — Export knowledge to Markdown, Obsidian, Notion, and exchange formats`
16	- OAuth authentication — Built-in `planopticon auth` for Google, Dropbox, Zoom, Notion, GitHub, and Microsoft
17	`- Smart frame extraction — Change detection for transitions + periodic capture (every 30s) for slow-evolving content like document scrolling`
18	`- People frame filtering — OpenCV face detection removes webcam/video conference frames, keeping only shared content (slides, documents, screen shares)`
19	`- Diagram extraction — Vision model-based classification detects flowcharts, architecture diagrams, charts, and whiteboards`
20	`- Knowledge graphs — Extracts entities and relationships, builds and merges knowledge graphs across videos`
21	`- Action item detection — Finds commitments, tasks, and follow-ups with assignees and deadlines`
22	`- Batch processing — Process entire folders of videos with merged knowledge graphs and cross-referencing`
23	`- Rich output — Markdown, HTML, PDF, Mermaid diagrams, SVG/PNG renderings, JSON manifests`
24	`- Cloud sources — Fetch videos from Google Drive, Dropbox, and many more cloud platforms`
25	`- Checkpoint/resume — Pipeline resumes from where it left off if interrupted — no wasted work`
26	`- Screengrab fallback — When extraction isn't perfect, captures frames with captions — something is always better than nothing`
27
28	`## Quick Start`
29
30	```bash
31	`# Install from PyPI`
32	`pip install planopticon`
33
34	`# Analyze a single video`
35	`planopticon analyze -i meeting.mp4 -o ./output`
36
37	`# Ingest documents and build a knowledge graph`
38	`planopticon ingest ./notes/ --output ./kb --recursive`
39
40	`# Chat with your knowledge base`
41	`planopticon companion --kb ./kb`
42
43	`# Run the planning agent interactively`
44	`planopticon agent --kb ./kb --interactive`
45
46	`# Query the knowledge graph`
47	`planopticon query stats`
48
49	`# Export to Obsidian`
50	`planopticon export obsidian --input ./kb --output ./vault`
51
52	`# Process a folder of videos`
53	`planopticon batch -i ./recordings -o ./output --title "Weekly Meetings"`
54
55	`# See available AI models`
56	`planopticon list-models`
57	```
58
59	`## Installation`
60
61	`=== "PyPI (Recommended)"`
62
63	```bash
64	`pip install planopticon`
65	```
66
67	`=== "With cloud sources"`
68
69	```bash
70	`pip install planopticon[cloud]`
71	```
72
73	`=== "With everything"`
74
75	```bash
76	`pip install planopticon[all]`
77	```
78
79	`=== "Binary (no Python needed)"`
80
81	`Download the latest binary for your platform from`
82	`[GitHub Releases](https://github.com/ConflictHQ/PlanOpticon/releases).`
83
84	`## Requirements`
85
86	`- Python 3.10+`
87	`- FFmpeg (for audio extraction)`
88	- At least one API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `GEMINI_API_KEY`
89
90	`## License`
91
92	`MIT License — Copyright (c) 2026 CONFLICT LLC. All rights reserved.`
93

PlanOpticon

Keyboard Shortcuts