Monitor real-time multi-agent activity and tool usage via WebSocket streaming.
npx skills add https://github.com/CarbeneAI/Forge --skill observability-carbeneai
Please help me install this Agent Skill. Skill: Observability Source: https://github.com/CarbeneAI/Forge/tree/main/.claude/skills/Observability Command: npx skills add https://github.com/CarbeneAI/Forge --skill observability-carbeneai
Real-time monitoring of PAI multi-agent activity and interactions to help you debug and understand agent workflows as they happen.
Start the observability dashboard using the provided control script: ~/.claude/skills/observability/manage.sh start
High-intent search queries and answers about installing and using this skill.
You can monitor real-time multi-agent activity by streaming hook events and tool calls to a live dashboard via WebSocket, ingesting daily JSONL event logs in-memory for immediate workflow debugging insights.
Debugging multi-agent workflows in real-time requires streaming agent sessions and hook events to a live UI, allowing you to trace interactions and monitor agent health instantly as they occur.
Start the observability dashboard by running the provided control script manage.sh with the start command, which initializes WebSocket streaming and in-memory event ingestion for your agents.
No persistent database is required; the dashboard uses in-memory ingestion from daily JSONL event logs with minimal persistence to stream agent activity and hook events efficiently.
Yes, you can trace hook events across multiple AI agents simultaneously by applying WebSocket streaming to a live dashboard, capturing tool usage and interactions for end-to-end operational monitoring.
rafaelcallejaStream Claude Code agent events to a filesystem-based observability dashboard.
fysoul17Map ingested sources to DOCX/HWPX templates and export filled documents.
danielmiesslerStream PAI multi-agent activity to a WebSocket-enabled dashboard.
zmreVisualize Claude Code agent interactions from JSONL logs via WebSocket streaming.
