v2.4
AI Multi-Agent Orchestration Hub
AI Multi-Agent Orchestration Hub
Inspect task flow, cross-agent statuses, and defect revision loops.
Execution sequence and dependency relationships across connected AI agents.
Live monitoring of tokens, rate windows, and session quotas across Claude, ChatGPT / Codex, Gemini, DeepSeek, and local harnesses.
ESC
Copy and paste this into Claude.ai, ChatGPT.com, or Gemini. Contains upstream outputs and instructions.
Ask upstream agent to fix defects or update specifications.
Marking this task complete unblocks whatever depends on it.
Every task downstream of this one is marked blocked, and the job is marked failed.
Leave blank to use default cloud endpoint.
Stored locally in encrypted SQLite database for agent dispatch.
For Web sessions: Open browser DevTools (F12) → Console, and run the 1-click extractor script.
Pair your iPhone or iPad to monitor live pipelines, dispatch commands, and track quotas.
Anyone with this token and network access has full control over your local machine and agents. Keep it confidential and pair only trusted devices.
Connections from outside your network use iroh's free public relays, run by a third party. Most connections are direct and never touch a relay. Relayed connections are rate limited and carry no uptime guarantee.
Scan with AgnView companion app on your phone or tablet to pair instantly.
How this session reaches the hub
Loading transport state...
Route remote connections around the public relays
Leave empty to use the free public relays. Paste your own relay URL to route around them.
Background orchestration and system boot settings
Launches AgnView server automatically in the background when your computer starts.
Automatically sync active web sessions from Claude.ai, ChatGPT, and Gemini directly into AgnView without manual copy-pasting.
Pipeline completion and failure webhooks
Scanning working directory...