跳到正文
cccskills
免费在 GitHub 公开

tmux-processes

Patterns for running long-lived processes in tmux. Use when starting dev servers, watchers, tilt, or any process expected to outlive the conversation.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(2)

  • SKILL.md5.8 KB
  • skill-report.json64.5 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

tmux Process Management

Interactive Shell Requirement

Use send-keys pattern for reliable shell initialization. Creating a session spawns an interactive shell automatically. Use send-keys to run commands within that shell, ensuring PATH, direnv, and other initialization runs properly.

# WRONG - inline command bypasses shell init, breaks PATH/direnv
tmux new-session -d -s "$SESSION" -n main 'tilt up'

# CORRECT - create session, then send command to interactive shell
tmux new-session -d -s "$SESSION" -n main
tmux send-keys -t "$SESSION:main" 'tilt up' Enter

Session Naming Convention

Always derive session name from the project:

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

For multiple processes in one project, use windows not separate sessions:

  • Session: myapp
  • Windows: server, tests, logs

Starting Processes

Single Process

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Create session with named window, then send command
tmux new-session -d -s "$SESSION" -n main
tmux send-keys -t "$SESSION:main" '<command>' Enter

Idempotent Start

Check if already running before starting:

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

if ! tmux has-session -t "$SESSION" 2>/dev/null; then
  tmux new-session -d -s "$SESSION" -n main
  tmux send-keys -t "$SESSION:main" '<command>' Enter
else
  echo "Session $SESSION already exists"
fi

Adding Windows to Existing Session

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Add a new window if it doesn't exist
if ! tmux list-windows -t "$SESSION" -F '#{window_name}' | grep -q "^server$"; then
  tmux new-window -t "$SESSION" -n server
  tmux send-keys -t "$SESSION:server" 'npm run dev' Enter
else
  echo "Window 'server' already exists"
fi

Multiple Processes (Windows)

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Create session with first process
tmux new-session -d -s "$SESSION" -n server
tmux send-keys -t "$SESSION:server" 'npm run dev' Enter

# Add more windows
tmux new-window -t "$SESSION" -n tests
tmux send-keys -t "$SESSION:tests" 'npm run test:watch' Enter

tmux new-window -t "$SESSION" -n logs
tmux send-keys -t "$SESSION:logs" 'tail -f logs/app.log' Enter

Monitoring Output

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Last 50 lines from first window
tmux capture-pane -p -t "$SESSION" -S -50

# From specific window
tmux capture-pane -p -t "$SESSION:server" -S -50

# Check for errors
tmux capture-pane -p -t "$SESSION" -S -100 | rg -i "error|fail|exception"

# Check for ready indicators
tmux capture-pane -p -t "$SESSION:server" -S -50 | rg -i "listening|ready|started"

Lifecycle Management

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# List all sessions (see what exists)
tmux ls

# List windows in current session
tmux list-windows -t "$SESSION"

# Kill only this project's session
tmux kill-session -t "$SESSION"

# Kill specific window
tmux kill-window -t "$SESSION:tests"

# Send keys to a window (e.g., Ctrl+C to stop)
tmux send-keys -t "$SESSION:server" C-c

Isolation Rules

  • Never use tmux kill-server
  • Never kill sessions not matching current project
  • Always derive session name from git root or pwd
  • Always verify session name before kill operations
  • Other Claude Code instances may have their own sessions running

When to Use tmux

ScenarioUse tmux?
tilt upYes, always
Dev server (npm run dev, rails s)Yes
File watcher (npm run watch)Yes
Test watcher (npm run test:watch)Yes
Database serverYes
One-shot build (npm run build)No
Quick command (<10s)No
Need stdout directly in conversationNo

Checking Process Status

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Check session exists
tmux has-session -t "$SESSION" 2>/dev/null && echo "session exists" || echo "no session"

# List windows and their status
tmux list-windows -t "$SESSION" -F '#{window_name}: #{pane_current_command}'

# Check if specific window exists
tmux list-windows -t "$SESSION" -F '#{window_name}' | grep -q "^server$" && echo "server window exists"

Restarting a Process

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Send Ctrl+C then restart command
tmux send-keys -t "$SESSION:server" C-c
sleep 1
tmux send-keys -t "$SESSION:server" 'npm run dev' Enter

Common Patterns

Start dev server if not running

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

if ! tmux has-session -t "$SESSION" 2>/dev/null; then
  tmux new-session -d -s "$SESSION" -n server
  tmux send-keys -t "$SESSION:server" 'npm run dev' Enter
  echo "Started dev server in tmux session: $SESSION"
elif ! tmux list-windows -t "$SESSION" -F '#{window_name}' | grep -q "^server$"; then
  tmux new-window -t "$SESSION" -n server
  tmux send-keys -t "$SESSION:server" 'npm run dev' Enter
  echo "Added server window to session: $SESSION"
else
  echo "Server already running in session: $SESSION"
fi

Wait for server ready

SESSION=$(basename $(git rev-parse --show-toplevel 2>/dev/null) || basename $PWD)

# Poll for ready message
for i in {1..30}; do
  if tmux capture-pane -p -t "$SESSION:server" -S -20 | rg -q "listening|ready"; then
    echo "Server ready"
    break
  fi
  sleep 1
done

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Stop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.

中文简介正在准备中。当前显示原文说明。

aiskillstore/marketplace4432026年10月23日 更新

Trigger for: explaining current FAR, DFARS, or agency-supplement text; determining documented acquisition-policy status for an agency and FAR part; comparing codified text, RFO model text, and agency deviations; tracing acquisition rulemaking; finding procurement comment periods; analyzing public comments; refreshing prior policy analysis; or producing an Acquisition Policy Impact Brief. Use published federal sources and supplied policy documents without making legal or procurement-specific applicability determinations. Do not use for clause selection, market research execution, opportunities, pricing, grants, or cooperative agreements.

中文简介正在准备中。当前显示原文说明。

aiskillstore/marketplace4432026年10月23日 更新

Systematic framework for extracting actionable items from documents and organizing them into prioritized, trackable checklists. Use when converting reports, meeting notes, audits, or any document with embedded action items into structured TODO lists.

中文简介正在准备中。当前显示原文说明。

aiskillstore/marketplace4432026年10月23日 更新

Browser automation for AI agents via inference.sh. Navigate web pages, interact with elements using @e refs, take screenshots, record video. Capabilities: web scraping, form filling, clicking, typing, drag-drop, file upload, JavaScript execution. Use for: web automation, data extraction, testing, agent browsing, research. Triggers: browser, web automation, scrape, navigate, click, fill form, screenshot, browse web, playwright, headless browser, web agent, surf internet, record video

中文简介正在准备中。当前显示原文说明。

aiskillstore/marketplace4432026年10月23日 更新

This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.

中文简介正在准备中。当前显示原文说明。

aiskillstore/marketplace4432026年10月23日 更新

Run AI apps via inference.sh CLI - image generation, video creation, LLMs, search, 3D, Twitter automation. Models: FLUX, Veo, Gemini, Grok, Claude, Seedance, OmniHuman, Tavily, Exa, OpenRouter, and many more. Use when running AI apps, generating images/videos, calling LLMs, web search, or automating Twitter. Triggers: inference.sh, infsh, ai model, run ai, serverless ai, ai api, flux, veo, claude api, image generation, video generation, openrouter, tavily, exa search, twitter api, grok

中文简介正在准备中。当前显示原文说明。

aiskillstore/marketplace4432026年10月23日 更新

查看 aiskillstore 的全部技能

报告此技能的问题