terminal-recorder-workflow

Capture tmux sessions into replayable artifacts and analyze them into parser/state views.

18|3|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/igamenovoer/houmao --skill terminal-recorder-workflow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: terminal-recorder-workflow
Source: https://github.com/igamenovoer/houmao/tree/main/context/skills/terminal-recorder-workflow
Command: npx skills add https://github.com/igamenovoer/houmao --skill terminal-recorder-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It solves the problem of auditing and debugging how a tmux-backed AI agent session behaves in practice by turning live terminal activity into replayable artifacts with parser and state-tracking outputs.

Core Features & Use Cases

  • Replay-grade capture: Records pane snapshots as the authoritative replay surface for later inspection.
  • Investigation and validation: Produces analyzed parser/state observations to compare against raw terminal snapshots when tracking logic or lifecycle goes wrong.
  • Operator labeling: Persists durable, scenario- and sample-level notes into labels.json so you can document observed terminal signals and checkpoints for future reference.
  • Use Case: Debug a suspected TUI lifecycle or parser misread by recording the exact tmux pane sequence, running analyze, and labeling the specific sample(s) where the observed state diverges from expectations.

Quick Start

Run the recorder manually with the repository CLI from the repo root after you have identified the exact tmux session and pane.

Frequently Asked Questions about terminal-recorder-workflow

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I record a tmux session to debug an AI agent's terminal behavior?

To debug tmux agent activity, you manually capture pane snapshots into replayable artifacts and then analyze them into parser and state-tracking views. This records the exact terminal sequence for later inspection.

What is the best way to track TUI lifecycle state during an automated agent workflow?

Tracking TUI lifecycle requires recording pane snapshots as the authoritative replay surface and running an analysis to generate state observations. You can label specific samples where observed state diverges from expectations.

Can I use terminal recording to investigate parser misreads in a tmux-backed agent?

Yes, terminal recording captures raw pane snapshots to investigate parser misreads. You must run the analyze step to produce parser observations before trusting the tracking logic against the original session.

Do I need to run an analysis before trusting the state tracking output of a terminal recording?

Yes, you must run analyze before trusting state tracking output. The analysis generates parser and state observations from the machine replay source of truth, ensuring accurate debugging of the recorded session.

How can I label observed terminal signals and checkpoints for future debugging reference?

You label observed terminal signals and checkpoints by persisting durable, scenario-level and sample-level notes into labels.json. This documents specific terminal behaviors for future reference during debugging.

What format serves as the machine replay source of truth for tmux terminal recording?

The pane_snapshots.ndjson file serves as the machine replay source of truth, while session.cast is used only for human review. This strict artifact contract ensures reliable debugging analysis.