Monitor Claude Code sub-agent activity with a TypeScript CLI.
npx skills add https://github.com/victorstein/claude-canvas --skill agent-monitor
Please help me install this Agent Skill. Skill: agent-monitor Source: https://github.com/victorstein/claude-canvas/tree/main/canvas/skills/agent-monitor Command: npx skills add https://github.com/victorstein/claude-canvas --skill agent-monitor
Real-time visibility into Claude Code sub-agent activity to simplify debugging, auditing, and coordination within a tabbed terminal interface.
Show the agent monitor in the current terminal: bun run src/cli.ts show agent-monitor --scenario monitor Spawn the agent monitor in a new WezTerm/tmux pane: bun run src/cli.ts spawn agent-monitor --scenario monitor With initial configuration: bun run src/cli.ts spawn agent-monitor --scenario monitor --config '{"parentSessionId": "session-123","agents":{}}'
High-intent search queries and answers about installing and using this skill.
You can monitor Claude Code sub-agent activity in real time by spawning a tabbed terminal interface that tracks per-agent state, output streams, and tool calls. This provides live visibility to support debugging and auditing concurrent agent workflows.
Tracking tool calls and git diffs across sub-agents requires a monitor with per-session state tracking and a tabbed interface. Each tab displays scrollable output streams, tool call details, and git diff visualizations for individual agents.
Yes, you can auto-spawn a terminal monitor when a sub-agent starts using hook-based auto-spawning. This automatically launches the monitoring interface in a new WezTerm or tmux pane whenever sub-agent activity begins.
Yes, the monitor supports multiple concurrent sub-agent sessions through per-session state tracking. You can configure initial sessions by passing a JSON config with a parent session ID and agent mapping when starting the CLI.
You start the sub-agent monitor in your current terminal by running the TypeScript CLI show command with the monitor scenario. This displays the tabbed monitoring interface directly in your existing terminal session.
Monitoring Claude Code sub-agent workflows relies on a TypeScript CLI and hook-based auto-spawning, requiring a compatible terminal environment like WezTerm or tmux. It is designed for safe operation but depends on terminal UI support for tabbed interfaces.
KunCheng-HeCoordinate terminal-based coding agents and background processes via a multiplexer interface.
JNZaderAnalyze per-sub-agent token usage and performance from Claude Code JSONL logs.
DicklesworthstoneManage WezTerm terminal panes, tabs, and workspaces via CLI.
tylerjrbuellConfigure @reactive-agents observability with verbosity levels, live streaming, JSONL export, model I/O logging, and audit trails for agent execution.
