CLI power tools for AI-assisted development. Use when (1) needing recommendations for CLI tools to install, (2) processing JSON/YAML data with jq/yq, (3) searching code with ripgrep or ast-grep, (4) documenting a CLI tool or multi-tool recipe you've discovered, (5) wanting to learn CLI patterns for data pipelines, or (6) setting up a new project and want CLI recommendations. Supports three modes - init (project scan), document (capture new recipes), and recommend (codebase analysis).
/plugin marketplace add killerapp/mem8-plugin/plugin install mem8@mem8-officialThis skill inherits all available tools. When active, it can use any tool Claude has access to.
clis/_index.mdclis/ast-grep/quick.mdclis/fd/quick.mdclis/ffmpeg/quick.mdclis/gh/quick.mdclis/jq/quick.mdclis/rg/quick.mdrecipes/CONTRIBUTING.mdrecipes/TEMPLATE.mdrecipes/_index.mdrecipes/data-processing/jq-deep-merge.mdrecipes/data-processing/rg-json-output.mdrecipes/devops/docker-cleanup-all.mdrecipes/devops/docker-stats-json.mdrecipes/github/gh-batch-issues.mdrecipes/github/gh-extract-issue-urls.mdrecipes/github/gh-pr-files-changed.mdrecipes/media/ffmpeg-compress-video.mdrecipes/media/ffmpeg-extract-thumbnail.mdreferences/install-guides.mdTransform terminal workflows with modern CLI utilities optimized for AI-assisted development.
clis/)Per-tool documentation with progressive disclosure:
recipes/)Multi-CLI combinations (2+ tools) for specific use cases:
Categories: github, data-processing, code-analysis, devops, media
When: Setting up new project, after /init
Purpose: Scan project, recommend relevant CLI patterns
python scripts/scan.py to detect CLIs in project configspython scripts/discover.py --missing to find missing Tier 1 toolsrecipes/ matching detected workflowsWhen: User discovers useful CLI pattern or recipe Purpose: Help capture terse, high-quality documentation
For single CLI:
{cli} --help, optionally WebSearch for docsclis/{cli}/quick.md following templateFor recipe (multi-CLI):
recipes/TEMPLATE.mdpython scripts/validate.pyWhen: User explicitly asks for CLI recommendations Purpose: Analyze codebase patterns, recommend tools
python scripts/discover.py for current tool statepython scripts/scan.py for project patternspython scripts/discover.py # Full report
python scripts/discover.py --tier 1 # Essentials only
python scripts/discover.py --missing # What to install
| Tool | Check | Purpose |
|---|---|---|
| jq | jq --version | JSON processing |
| ripgrep | rg --version | Fast code search |
| fd | fd --version | File finder |
| bat | bat --version | Cat with syntax |
| ast-grep | sg --version | AST refactoring |
# Quick check all Tier 1
for cmd in jq rg fd bat sg; do command -v $cmd &>/dev/null && echo "OK $cmd" || echo "MISSING $cmd"; done
# macOS
brew install jq ripgrep fd bat ast-grep
# Ubuntu/Debian
sudo apt install jq ripgrep fd-find bat
cargo install ast-grep --locked
# Windows (Scoop)
scoop install jq ripgrep fd bat ast-grep
jq '.data[].name' response.json # Extract field
jq '.items[] | select(.active)' data.json # Filter
jq -r '.users[] | [.id, .name] | @csv' users.json # To CSV
jq -s '.[0] * .[1]' base.json override.json # Merge
rg 'TODO|FIXME' -A 2 --type py # With context
rg 'pattern' --json | jq 'select(.type=="match")' # Structured output
rg -l 'deprecated' | xargs sd -i 'old' 'new' # Search & replace
gh issue view 123 --json body --jq '.body' # Inline jq
gh pr view 123 --json files --jq '.files[].path' # PR files
gh issue list --json number,title --jq '.[].title' # List issues
When user asks about CLI tools:
clis/{tool}/quick.md for specific tool questionsclis/{tool}/reference.md only for advanced usagerecipes/{category}/*.md for multi-tool workflowsCreate ~/.claude/skills/cli-ninja-local/:
cli-ninja-local/
├── SKILL.md # References this skill
├── recipes/ # Personal recipes
└── clis/ # Personal CLI docs
recipes/TEMPLATE.mdpython scripts/validate.py your-recipe.mdSee recipes/CONTRIBUTING.md for details.
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
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.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.