rtk-ai/rtk

Health v2 · maintenance only

Adoption not part of the score

81836 stars · 5190 forks observed · 2026-09-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-21, confidence 0.99

RTK (Rust Token Killer) is a high‑performance command‑line proxy that compresses and filters shell command output before it reaches a language model. It is delivered as a single Rust binary with zero runtime dependencies.

Use cases

• reduce LLM token usage when running development commands in an AI coding assistant
• compress git diff or git log output for more efficient context in code review bots
• filter test runner output (cargo test, pytest) to only include failures for AI debugging
• minimize noise from ls, grep, or find commands for AI agents that process file system data
• lower costs of using paid LLM APIs by cutting unnecessary CLI output

When to choose

• You need a drop‑in tool to shrink CLI output for any LLM‑powered developer assistant
• You prefer a single, statically compiled binary with no external dependencies
• Your workflow runs many repetitive commands whose full output overwhelms the model's context window

When to avoid

• You require full, unaltered command output for downstream processing or auditing
• You need support for commands not covered by RTK’s built‑in filters
• You are using a language model with an extremely large context window where token savings are negligible

Facets

cli-tool · maturity active

cli middleware artificial-intelligence developer-tools windows cli rust token-saver llm-cost-reduction shell-output-compression linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="rtk-ai/rtk")