jfrog/boost

What it solves

Boost reduces the token cost and noise associated with AI coding agents (like Cursor, Claude Code, and GitHub Copilot) by compressing the verbose output of shell commands. It prevents agents from wasting context window space on repetitive logs, such as dependency installations or successful build steps, while ensuring critical information like errors and timings are preserved.

How it works

Boost acts as a wrapper for shell commands. Instead of passing raw, noisy logs directly to the agent, it applies command-aware filters to trim unnecessary data. It preserves essential signals—such as failure messages, compiler errors, and stack frames—and provides compact summaries for successful operations. It also features a local cache to further reduce output for repeated tasks and supports custom TOML filters for internal tools.

Who it’s for

• AI Agent Users: Developers using coding agents who want to lower token usage and keep the agent's context lean during long sessions.
• DevOps/CI Engineers: Those looking for shorter, more scannable job logs in CI pipelines (e.g., GitHub Actions).
• Tool Developers: People who want to add structured compression to their own custom CLI tools.

Highlights

• Command-Aware Compression: Specifically targets noise in tools like npm, pytest, docker, and git without losing critical error data.
• Agent Integration: Designed to wire directly into Cursor, Claude Code, GitHub Copilot, and Codex CLI.
• Performance Validated: Benchmarked to maintain identical task pass rates while reducing costs by approximately 12%.
• Observability: Includes an interactive web report and terminal-based summaries to track token savings.
• Enterprise Ready: OpenTelemetry-enabled and features local-first privacy with automatic secret redaction.