skills.rest
Base-Analitica
Base-AnaliticaOFFICIAL
Explore all skills by this author →
GitHub

token-efficient-shell

Compress verbose shell command output using the project-local RTK wrapper.

Updated Aug 19, 2026
Search Tags:#git#context management#pytest#token efficiency#shell#rtk#output compression
One-click install
npx skills add https://github.com/Base-Analitica/mouse-hub --skill token-efficient-shell-base-analitica
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: token-efficient-shell
Source: https://github.com/Base-Analitica/mouse-hub/tree/main/.jcode/skills/token-efficient-shell
Command: npx skills add https://github.com/Base-Analitica/mouse-hub --skill token-efficient-shell-base-analitica

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Verbose output from commands like git status, pytest, grep, and find consumes excessive context tokens when working with an AI assistant, leaving less room for actual reasoning and code. ## Core Features & Use Cases - Output Compression: Routes supported verbose commands (git, rg, grep, find, ls, tree, pytest, linters, builds) through the project-local scripts/agent/rtk wrapper to produce compact summaries. - Raw Output Preservation: Keeps exact, unfiltered output for byte-sensitive data, patches, and protocol output, and stores raw failure output under .tools/rtk/state/tee/ for later inspection. - Savings Reporting: The scripts/agent/rtk gain command shows local token savings from compressed output. - Use Case: While debugging a failing test suite, run pytest through RTK to get a compact failure summary in context, then inspect the tee path for the full raw traceback only when needed. ## Quick Start Run the bootstrap script once per checkout with scripts/agent/bootstrap-rtk, then invoke supported shell commands through scripts/agent/rtk instead of the native command.

Frequently Asked Questions about token-efficient-shell

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

FAQPage Schema
How do I reduce token usage from shell command output?▼

Run supported verbose commands through the project-local RTK wrapper at scripts/agent/rtk instead of the native command. It produces compact summaries for git, grep, find, pytest, linters, and builds while keeping raw output available.

Which commands does RTK support for output compression?▼

RTK supports git status, diff, and log, plus rg, grep, find, ls, tree, pytest, linters, and build commands. For anything else, run the native command directly.

When should I avoid using RTK output filtering?▼

Use the native command for exact patches, byte-sensitive data, protocol output, or small results where compression adds no value. Do not stack another output filter on top of RTK.

How do I see the full raw output after an RTK failure?▼

RTK prints a tee path when a command fails. Inspect that path to read the raw failure output, which is stored under .tools/rtk/state/tee/ before rerunning the command.

What happens if RTK is not installed in my checkout?▼

RTK is optional. If it is unavailable, run the native command directly. To set it up, run scripts/agent/bootstrap-rtk once per checkout.

Related Skills

Explore More Software Engineering Skills→
wesleysimpliciowesleysimplicio
1

rtk-cli

Compress verbose shell output before it becomes AI agent context.

Community
Intermediate
udecodeudecode
1.2k

systematic-debugging

Diagnose root causes of software bugs through a four-phase investigation workflow.

Official
Intermediate
marinvchmarinvch

context-fundamentals

Optimize AI agent context budgets with modular prompts and token estimation.

Community
Advanced
parisgroup-aiparisgroup-ai
1

railway-logs

Fetch and analyze Railway production logs to diagnose deployment and runtime issues.

Official
Intermediate