using-tmux-for-interactive-commands
Use when you need to run interactive CLI tools (vim, git rebase -i, Python REPL, etc.) that require real-time input/output - provides tmux-based approach for controlling interactive sessions through detached sessions and send-keys
Third-Party Agent Skill: Review the code before installing. Agent skills execute in your AI assistant's environment and can access your files. Learn more about security
Installation for Agentic Skill
View all platforms →skilz install obra/superpowers-lab/using-tmux-for-interactive-commands skilz install obra/superpowers-lab/using-tmux-for-interactive-commands --agent opencode skilz install obra/superpowers-lab/using-tmux-for-interactive-commands --agent codex skilz install obra/superpowers-lab/using-tmux-for-interactive-commands --agent gemini
First time? Install Skilz: pip install skilz
Works with 22+ AI coding agents
Cursor, Aider, Copilot, Windsurf, Qwen, Kimi, and more...
Extract and copy to ~/.claude/skills/ then restart Claude Desktop
git clone https://github.com/obra/superpowers-lab cp -r superpowers-lab/skills/using-tmux-for-interactive-commands ~/.claude/skills/ Need detailed installation help? Check our platform-specific guides:
Related Agentic Skills
claude-permissions
by SpillwaveSolutionsProactive Claude Code permission manager. Configures permissions via natural language for CLI tools (git, gcloud, aws, kubectl, maven, gradle, npm, do...
developing-llamaindex-systems
by SpillwaveSolutionsProduction-grade agentic system development with LlamaIndex in Python. Covers semantic ingestion (SemanticSplitterNodeParser, CodeSplitter, IngestionP...
mastering-langgraph
by SpillwaveSolutionsBuild stateful AI agents and agentic workflows with LangGraph in Python. Covers tool-using agents with LLM-tool loops, branching workflows, conversati...
mastering-langgraph
by SpillwaveSolutionsBuild stateful AI agents and agentic workflows with LangGraph in Python. Covers tool-using agents with LLM-tool loops, branching workflows, conversati...
Agentic Skill Details
- Repository
- superpowers-lab
- Type
- Technical
- Meta-Domain
- development
- Primary Domain
- python
- Market Score
- 74.0
Agent Skill Grade
C
Score: 74/100
Click to see breakdown
Score Breakdown
Areas to Improve
- All 179 lines in single file instead of concise overview with references directory
- 179-line file exceeds 100-line threshold but lacks table of contents
- References /home/jesse/git/interactive-command/tmux-wrapper.sh which won't exist for users
Recommendations
- Address 1 high-severity issues first
- Add trigger phrases to description for discoverability
- Add table of contents for files over 100 lines
Graded: 1/5/2026
Developer Feedback
I took a look at your using-tmux-for-interactive-commands skill and wanted to share some thoughts.
Links:
The TL;DR
You're at 74/100, solid C-grade territory. This skill is based on Anthropic's best practices for agentic skills. Your strongest area is Ease of Use (21/25)—the metadata and trigger terms are spot-on. The weakest link is Progressive Disclosure Architecture (19/30)—everything's crammed into one 179-line file when it should be layered.
What's Working Well
- Excellent trigger phrases in the description (vim, git rebase -i, Python REPL). That specificity is exactly what makes skills discoverable.
- Practical examples with real I/O—the vim and Python REPL examples show exactly what the commands do. That's chef's kiss for clarity.
- Solid utility core. You're solving a real gap: Claude can't naturally control interactive tools. That's genuinely valuable for agent workflows.
- Consistent terminology throughout (session, send-keys, capture-pane). Makes it way easier to follow.
The Big One: Single File Should Be Layered
Right now you've got 179 lines in one SKILL.md. That kills your PDA score (19/30) because users have to scroll through everything to find what they need.
The fix: Create a references/ directory and split the skill:
using-tmux-for-interactive-commands/
├── SKILL.md (80 lines max - overview + quick reference + Core Pattern)
├── references/
│ ├── common-patterns.md (detailed patterns)
│ ├── common-mistakes.md (troubleshooting)
│ └── tmux-wrapper.sh (bundled script)
Keep SKILL.md focused on the essentials, use frontmatter links to pull in references when needed. This nets you +4 points easily.
Other Things Worth Fixing
Hardcoded external path (line 90):
/home/jesse/git/interactive-command/tmux-wrapper.shwon't exist for anyone else. Change it to reference a bundled./tmux-wrapper.shin the skill directory. (+1 point)Missing table of contents for a 179-line file. Add one after the frontmatter—at this length it helps users jump to sections. (+1 point)
Second-person phrasing in description: "Use when you need to run..." should be "Use for running..." Removes the implied "you" and tightens it up.
Real-World Impact section reads like marketing copy ("Enables programmatic control", "Unblocks any tool"). Either remove it or convert to a factual capability list.
Quick Wins
- Highest impact: Restructure with references/ directory (split that 179-line file)
- Next: Bundle the wrapper script and fix that hardcoded path
- Polish: Add a TOC, drop the second-person phrasing, tone down the marketing language
These changes would push you from 74 to ~82, solidly into B territory.
Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.
Browse Category
More development Agentic SkillsReport Security Issue
Found a security vulnerability in this agent skill?