Learning HubACFS Academy
Progress
0%
0 of 3333 remaining
  • 1
    Welcome & Overview5 min
  • 2
    Linux Navigation8 min
  • 3
    SSH & Persistence6 min
  • 4
    tmux Basics7 min
  • 5
    Git Essentials10 min
  • 6
    GitHub CLI8 min
  • 7
    Agent Commands10 min
  • 8
    NTM Command Center8 min
  • 9
    NTM Prompt Palette6 min
  • 10
    The Flywheel Loop10 min
  • 11
    Keeping Updated4 min
  • 12
    UBS: Code Quality Guardrails8 min
  • 13
    Agent Mail Coordination10 min
  • 14
    CASS: Learning from History8 min
  • 15
    The Memory System8 min
  • 16
    Beads: Issue Tracking8 min
  • 17
    Safety Tools: SLB & CAAM6 min
  • 18
    The Art of Agent Direction12 min
  • 19
    Case Study: cass-memory15 min
  • 20
    Case Study: SLB12 min
  • 21
    RU: Multi-Repo Mastery10 min
  • 22
    DCG: Pre-Execution Safety8 min
  • 23
    Meta Skill: Local Skills10 min
  • 24
    SRPS: System Protection8 min
  • 25
    JFP: Prompt Library6 min
  • 26
    APR: Automated Plan Reviser8 min
  • 27
    PT: Process Triage6 min
  • 28
    XF: X Archive Search6 min
  • 29
    RCH: Remote Compilation8 min
  • 30
    WA: WezTerm Automata8 min
    NOW
  • 31
    Brenner Bot: Research10 min
  • 32
    GIIL: Cloud Image Downloads6 min
  • 33
    S2P: Source to Prompt6 min
Back to Home
Back
30/33
Lesson 30
8 min

WA: WezTerm Automata

Real-time terminal observation and automation for AI agents

New to ACFS?

Complete the setup wizard first to get the most from these lessons.

Go to Choose Your OS
Goal

Coordinate multiple AI coding agents running in WezTerm with real-time observation and automation.

What Is WA?

WezTerm Automata (WA) is a terminal hypervisor that captures pane output in real-time, detects agent state transitions through pattern matching, and enables event-driven automation.

When running multiple AI agents, you need to know when they hit rate limits, complete tasks, or need approval. WA observes all panes with sub-50ms latency and can trigger automated responses.

Real-time Observation

Sub-50ms pane monitoring

Pattern Detection

Recognizes agent state changes

Full-Text Search

FTS5 with BM25 ranking

Safety Engine

Capability gates & audit trails

How It Works

WA runs as a daemon that observes all WezTerm panes, detects patterns, and exposes a Robot Mode API for agent integration.

Observation Flow
# WA observes terminal panes
Agent types: "Thinking..."
↓
# Pattern detection engine
↓
Agent completes: "Done!"
↓
# Event triggers automation
↓
WA notifies other agents via mail
Pro Tip
WA uses delta extraction instead of full buffer snapshots for minimal overhead.

Quick Start

Get started with these essential commands.

Start the observation daemon
Check daemon status
View all panes as JSON
Search captured output
Warning
WA requires WezTerm to be running. It won't work with other terminal emulators.

Robot Mode API

Robot Mode provides a JSON API for machine-to-machine communication.

Get all pane states as JSON
Get pane output
Send input to pane
Wait for pattern
Robot Mode Example
# Example: Wait for agent completion
wa robot wait-for 42 "Task complete"
# Then trigger next action
wa robot send 43 "Start next task"

Pattern Detection

WA detects state transitions for common AI agents.

Claude Code

Rate limits, approvals, completions

Codex CLI

Task completion, errors, waiting

Gemini CLI

Response completion, quota warnings

Custom Patterns

Define your own detection rules

Tool Integration

WA integrates seamlessly with other flywheel tools.

WA + NTM

WA automatically observes agents spawned by NTM. Use NTM to spawn agents and WA to monitor their state.

WA + Agent Mail

State changes detected by WA can trigger Agent Mail notifications. Coordinate agent handoffs through mail threads.

WA + Beads

When WA detects task completion, it can update bead status. Track agent progress through your issue tracker.

Diagnostics

Troubleshoot issues with built-in diagnostics.

Run diagnostic checks
Check daemon health
View recent logs
Note
Run wa doctor if pattern detection isn't working. It checks WezTerm connectivity and daemon status.

Ready to level up?

Mark complete to track your learning progress.

Previous
RCH: Remote Compilation
Next
Brenner Bot: Research