MCPInstall

rtk

by akillnessBackend & APIs

Install, initialize, verify, and troubleshoot RTK (Rust Token Killer) for AI coding agents. Use when you need to reduce shell-command token output, confirm that the correct rtk binary is installed, choose between Homebrew, install.sh, or Cargo installation, wire rtk init for Claude Code, Codex, Gemini CLI, Cursor, Copilot, Windsurf, Cline, or OpenCode, or use compact wrappers such as rtk git status, rtk read, rtk grep, rtk test, rtk lint, and rtk gain. Triggers on: rtk, rust token killer, token saver cli, rtk init, rtk gain, codex rtk, gemini rtk, opencode rtk, claude hook token reduction.

Installs
672
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/akillness/jeo-skills --skill rtk
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads rtk automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

rtk - Rust Token Killer for AI Coding Agents

Keyword: rtk · rust token killer · rtk init · rtk gain

The target repository is the akillness/rtk fork of rtk-ai/rtk. The fork tracks upstream code, while the bundled install.sh still downloads release artifacts from rtk-ai/rtk.

RTK is a high-performance CLI proxy that rewrites common shell commands into token-optimized equivalents before their output reaches an AI agent. Use this skill to verify the right RTK binary, install or repair it, choose the correct rtk init mode for the user's agent, and fall back to direct rtk commands when no automatic hook exists.

When to use this skill

  • Verify whether RTK is already installed and whether it is the correct project
  • Repair the common cargo install rtk name-collision mistake
  • Install RTK with Homebrew, the repository install.sh, or Cargo
  • Initialize RTK for Claude Code, Codex, Gemini CLI, Cursor, Copilot, Windsurf, Cline, or OpenCode
  • Explain when rtk init -g is better than local project setup
  • Use direct rtk wrappers such as rtk git status, rtk read, rtk grep, rtk test, or rtk lint
  • Diagnose why automatic rewriting is not happening
  • Explain why built-in file tools do not pass through shell hooks and when to use direct rtk commands instead

Instructions

Step 1: Verify the current installation before reinstalling anything

Run the local status helper first:

bash scripts/check-status.sh --show

Interpret the result in this order:

  1. If rtk gain works, the correct RTK is already installed. Do not reinstall unless the user wants an upgrade.
  2. If rtk --version works but rtk gain fails, the user likely has the wrong rtk package, usually Rust Type Kit.
  3. If rtk is missing entirely, move to installation.

The install and verification decision tree lives in references/install-and-verify.md.

Step 2: Install or repair RTK with the narrowest method that fits

Use the wrapper:

bash scripts/install.sh

Important installer choices:

  • Default auto prefers Homebrew on macOS when available, then the fork's install.sh, then Cargo
  • Use --method brew when Homebrew is available and the user wants a standard system-managed install
  • Use --method script for the repository install script path
  • Use --method cargo when the user explicitly wants a git-source build or is working from the fork
  • Use --force-uninstall-wrong when the user already installed the wrong rtk

Examples:

bash scripts/install.sh --method brew
bash scripts/install.sh --method script
bash scripts/install.sh --method cargo --repo akillness/rtk
bash scripts/install.sh --force-uninstall-wrong

Keep these installation rules explicit:

  • Always verify with rtk gain after installation
  • Treat cargo install rtk as unsafe unless the user already confirmed the correct package
  • If the install script path is used, remember the script still pulls release artifacts from rtk-ai/rtk

Step 3: Choose the correct rtk init mode for the agent

Use the init wrapper instead of hand-building flags:

bash scripts/init-agent.sh --agent claude --global

Default mapping:

  • Claude Code: --agent claude --global for hook-first setup
  • Codex: --agent codex --global or local --agent codex --local
  • Gemini CLI: --agent gemini --global
  • Cursor: --agent cursor --global
  • Copilot: --agent copilot --global
  • OpenCode: --agent opencode --global
  • Windsurf: --agent windsurf --local
  • Cline / Roo Code: --agent cline --local

Good follow-up checks:

bash scripts/init-agent.sh --agent claude --global --auto-patch
bash scripts/init-agent.sh --agent codex --global
bash scripts/init-agent.sh --agent opencode --global
bash scripts/init-agent.sh --agent claude --show

Read references/platform-init.md when the user needs the per-agent hook type, scope, or uninstall behavior.

Step 4: Use direct RTK commands when hooks are absent or insufficient

Automatic rewrite only applies where the agent actually executes shell commands through a supported hook or plugin. Prompt-only integrations such as Codex, Windsurf, and Cline still benefit from explicit rtk commands, but they do not transparently mutate built-in file tools.

Use direct commands when:

  • The agent uses a built-in Read, Grep, or Glob tool instead of shell
  • The workflow is in Codex and the user wants compact output immediately
  • The user wants deterministic verification independent of hook state

Common commands:

rtk git status
rtk read src/main.rs
rtk grep "pattern" .
rtk test cargo test
rtk lint
rtk gain

The command families and config knobs are summarized in references/commands-and-config.md.

Step 5: Diagnose failures explicitly

Use bash scripts/check-status.sh --show as the first diagnostic pass, then branch:

  • Wrong binary installed: fix the name collision and reinstall
  • Hook missing or stale: rerun the correct rtk init mode
  • Agent still verbose: confirm whether the task used shell commands or built-in tools
  • PATH issue: ensure ~/.local/bin or ~/.cargo/bin is on PATH
  • Windows wrapper failure: recommend RTK 0.23.1+

The troubleshooting matrix is in references/troubleshooting.md.

Examples

Example 1: Verify whether the correct RTK is already installed

bash scripts/check-status.sh --show

Example 2: Install from the fork-aware script path

bash scripts/install.sh --method script --repo akillness/rtk

Example 3: Repair the wrong cargo install rtk package

bash scripts/install.sh --force-uninstall-wrong --method cargo

Example 4: Claude Code global hook setup

bash scripts/init-agent.sh --agent claude --global --auto-patch

Example 5: Codex global prompt setup

bash scripts/init-agent.sh --agent codex --global

Example 6: OpenCode plugin install

bash scripts/init-agent.sh --agent opencode --global

Example 7: Use RTK directly for compact output

rtk read Cargo.toml
rtk grep "rtk init" README.md
rtk test cargo test

Example 8: Temporarily bypass rewrite for one command

RTK_DISABLED=1 git status

Best practices

  1. Check rtk gain before reinstalling; it is the quickest way to distinguish the correct RTK from the wrong package.
  2. Use the lightest install path that meets the user's needs: Homebrew for convenience, install script for release binaries, Cargo for source-driven workflows.
  3. Treat rtk init -g as the normal path for hook-capable agents and local rules files as the normal path for Windsurf and Cline.
  4. Use rtk init --show after setup or upgrades so stale hook state is visible immediately.
  5. Remember that built-in read/search tools do not pass through shell hooks; switch to shell commands or direct rtk wrappers when token savings matter.
  6. Keep RTK_DISABLED=1 in mind for one-off passthrough commands instead of uninstalling hooks.
  7. Re-run the installer plus rtk init after RTK upgrades because hook artifacts can change between releases.
  8. When users say "RTK is broken", separate installation correctness, hook registration, and actual command path before proposing a fix.

References

  • references/install-and-verify.md
  • references/platform-init.md
  • references/commands-and-config.md
  • references/troubleshooting.md
  • scripts/install.sh
  • scripts/init-agent.sh
  • scripts/check-status.sh
  • RTK Repository Fork
  • Upstream RTK Project

FAQ

What does the rtk skill do?

Install, initialize, verify, and troubleshoot RTK (Rust Token Killer) for AI coding agents. Use when you need to reduce shell-command token output, confirm that the correct rtk binary is installed, choose between Homebrew, install.sh, or Cargo installation, wire rtk init for Claude Code, Codex, Gemini CLI, Cursor, Copilot, Windsurf, Cline, or OpenCode, or use compact wrappers such as rtk git status, rtk read, rtk grep, rtk test, rtk lint, and rtk gain. Triggers on: rtk, rust token killer, token saver cli, rtk init, rtk gain, codex rtk, gemini rtk, opencode rtk, claude hook token reduction.

How do I install rtk?

Run npx skills add https://github.com/akillness/jeo-skills --skill rtk in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/rtk/ (personal) or .claude/skills/rtk/ (project).

Which AI agents can use rtk?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is rtk safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains rtk?

It is published by akillness in akillness/jeo-skills. It has 672 recorded installs.

More from akillness/jeo-skills

View all →

graphify

akillness/jeo-skills

Drive Graphify from its CLI to build, refresh, query, export, and serve a durable code/corpus knowledge graph. Use when the user wants .graphify/GRAPHREPORT.md, graph.json, graph.html, graphify update/summary/query/path/explain/tree, change-aware review context, git-hook or watch-based refresh, a stdio MCP graph server, or an install into jeo, jeopi, gjc, opencode, Claude, Codex, or Gemini. Also covers the honest structural fallback when native extraction is empty or misleading. Route simple locate/reference work to codebase-search, narrative knowledge-base work to llm-wiki, and project-memory handoff to opencontext. Triggers on: graphify, graphify update, graphify query, knowledge graph CLI, GRAPHREPORT.md, graph.json, codebase graph, graph refresh, graphify install, graphify serve, review context, affected flows.

Code Quality & Workflow662 installs

data-analysis

akillness/jeo-skills

Analyze datasets to extract insights, identify patterns, and generate reports.

598 installs

code-refactoring

akillness/jeo-skills

Turn a cleanup packet into one behavior-preserving refactor brief. Use when simplifying a messy function/component/service, freezing behavior before touching fragile legacy code, splitting a cleanup-heavy diff into reviewable slices, or planning a repeated migration / codemod without changing intended behavior. Route diagnosis to debugging, review judgment to code-review, validation-program design to testing-strategies, bottleneck-led tuning to performance-optimization, and pure symbol inventory to codebase-search.

Code Quality & Workflow546 installs

technical-writing

akillness/jeo-skills

Write internal technical documentation for engineers and operators: technical specs, architecture docs, ADRs, runbooks, migration plans, and developer-facing implementation guides. Use when the main job is capturing a technical decision, system boundary, operating procedure, or rollout path for builders and maintainers. Triggers on: tech spec, design doc, architecture doc, ADR, runbook, migration guide, implementation guide, rollout doc, operational guide, and internal technical writing. End-user onboarding guides, tutorials, FAQs, and help-center flows belong here too. Route API portals to api-documentation, release notes to changelog-maintenance, decks to presentation-builder, and GTM messaging to marketing-automation.

Docs & Writing544 installs

web-accessibility

akillness/jeo-skills

Routing-first skill for web accessibility remediation and verification. Use when the main job is classifying which accessibility surface is failing — semantics, keyboard/focus, labels/announcements, visual perception/reflow, media alternatives, or routed-app navigation feedback — and turning vague audit, WCAG, axe/Lighthouse, or "make this accessible" requests into one concrete remediation packet. Broad UI critique for hierarchy, polish, and launch readiness belongs here too. Route component API architecture and system governance to design-system, and responsive layout strategy to responsive-design.

Frontend & Design531 installs

remotion-video-production

akillness/jeo-skills

Compatibility alias for video-production when a prompt, setup surface, or legacy workflow explicitly asks for Remotion video production. Use when the request names Remotion, React-based video rendering, compositions, scene components, or a legacy remotion-video-production catalog entry, but the real workflow is still the canonical programmable-video lane. Prefer video-production for general use. Triggers on: remotion-video-production, Remotion render pipeline, React video composition, migrate old Remotion skill.

Media & Creative520 installs

Popular backend & apis skills

View category →

find-skills

vercel-labs/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

OfficialBackend & APIs4.1M installs

entra-app-registration

microsoft/azure-skills

Guides Microsoft Entra ID app registration, OAuth 2.0 authentication, and MSAL integration. USE FOR: create app registration, register Azure AD app, configure OAuth, set up authentication, add API permissions, generate service principal, MSAL example, console app auth, Entra ID setup, Azure AD authentication. DO NOT USE FOR: Key Vault secrets (use azure-keyvault-expiration-audit), general Azure resource security guidance.

OfficialBackend & APIs566.7K installs

google-agents-cli-adk-code

google/agents-cli

This skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use state management" — in a project that needs ADK (Agent Development Kit) API patterns and code examples. It provides a quick reference for agent types, tool definitions, orchestration patterns, callbacks, state management, the graph Workflow API, and reference recipes to study. Do NOT use for scaffolding (use google-agents-cli-scaffold) or deployment (use google-agents-cli-deploy).

Backend & APIs514.8K installs

embedded-captions

heygen-com/hyperframes

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet anchor rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.

Backend & APIs453.3K installs

entra-agent-id

microsoft/azure-skills

Provision Microsoft Entra Agent Identity Blueprints, BlueprintPrincipals, and per-instance Agent Identities via Microsoft Graph, and configure OAuth 2.0 token exchange (fmipath, OBO, cross-tenant) including the Microsoft Entra SDK for AgentID sidecar. USE FOR: Agent Identity Blueprint, BlueprintPrincipal, agent OAuth, fmipath token exchange, agent OBO, Workload Identity Federation for agents, polyglot agent auth, Microsoft.Identity.Web.AgentIdentities. DO NOT USE FOR: standard Entra app registration (use entra-app-registration), Microsoft Foundry agent authoring (use microsoft-foundry).

OfficialBackend & APIs336.6K installs

neon

neondatabase/agent-skills

Overview of Neon, a complete set of cloud backend primitives around Lakebase Postgres: Auth, Object Storage, Functions, the AI Gateway, and Realtime. Start here to choose Neon for undecided login, files, APIs, and LLM calls, set up the CLI or MCP server, and follow the branch-first workflow. Use when building an app or backend on Neon, or when "Neon" or "Lakebase Postgres" is mentioned. Child skill neon-postgres wins for an existing DATABASEURL, SQL, schema, inspect, or search. Child skill neon-auth wins for login, users, sessions, identity routing, and Managed Better Auth setup. Child skill neon-realtime wins for live queries in reactive apps and agents. Also use for object storage, S3, buckets, serverless functions, function triggers, cron, AI gateway, LLM calls, realtime, logs, Loki, Grafana, observability, postgres, database, backend, Claimable Neon, neon.new, or a no-signup database.

Backend & APIs237.1K installs