TE

terminal-context

Controls and monitors Kitty terminal sessions directly from your coding agent.

Install

mkdir -p .claude/skills/terminal-context && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/426" && unzip -o skill.zip -d .claude/skills/terminal-context && rm skill.zip

Installs to .claude/skills/terminal-context

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Complete Kitty terminal awareness + control for coding agents: list panes/tabs, read scrollback, map ports→processes, parse per-pane git/last-command metadata (from shell hooks), and send commands/focus panes. Use when user mentions "another terminal", "is the server running", "what failed", or you need to run/inspect commands across panes.
342 chars · catalog description✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Advanced

Key capabilities

  • List active Kitty terminal tabs and panes
  • Inspect scrollback buffer content across panes
  • Map open network ports to running processes
  • Parse per-pane shell command metadata
  • Send remote commands to specific terminal panes

How it works

It uses a local Kitty socket to query the window manager's state and inject commands into specific terminal pane instances.

Inputs & outputs

You give it
Commands or queries regarding terminal state or process status.
You get back
Process lists, error logs from buffers, and confirmation of command execution.

When to use terminal-context

  • Identify which terminal pane is running a dev server
  • Inspect error logs in a specific tab's scrollback
  • Execute tests across multiple terminal windows
  • Restart processes based on port conflicts

About this skill

terminal-context

Default rule

Prefer tc-context, tc-output, tc-send, and tc-watch over manual copy/paste when you need terminal state from other Kitty panes.

Quickstart

  1. tc-context --summary
  2. tc-context --insights
  3. tc-output --cwd "$PWD" --lines 120 --match "error|failed|traceback"
  4. tc-send --cwd "$PWD" "npm test"
  5. tc-watch --cwd "$PWD" "ready|listening|compiled" --timeout 90

Safety contract

  • Prefer read operations (tc-context, tc-output) before sending commands to other panes.
  • Before destructive actions, confirm you have the correct pane and cwd.

Setup

  • For deterministic/non-interactive use, enable a Kitty socket (listen_on) and install shell hooks. See references/SETUP.md.

When not to use it

  • Standard GUI terminal windows
  • Local machine without Kitty terminal

Prerequisites

Kitty terminalTc-context shell hooks

Limitations

  • Only compatible with Kitty terminal
  • Requires shell hooks for full metadata
  • Execution depends on socket permissions

How it compares

It provides programmatic control over the terminal environment rather than operating solely through manual user interaction.

Compared to similar skills

terminal-context side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
terminal-context (this skill)57moReviewAdvanced
resolve-conflicts818moReviewIntermediate
openspec-onboard106moReviewBeginner
codex-cli-bridge99moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry