How does the skill handle large volumes of error output?

The skill intelligently deduplicates repeating errors, prioritizes critical failures over warnings, and truncates the report to the top 10 most relevant issues.

Which programming runtimes does this skill support?

It supports a wide range of runtimes including Bun, NPM, PNPM, Yarn, Go, Rust (Cargo), and Python (UV, Poetry, or Pip).

What types of development tasks can I run?

It is optimized for standard development workflows including installation, building, testing, type-checking, linting, formatting, and cleaning.

Does it detect my project configuration automatically?

Yes, it uses a detection priority list looking for files like agentuity.json, lockfiles (bun.lockb, package-lock.json), and manifest files (go.mod, Cargo.toml) to determine the correct environment.

Agentuity Command Runner

Name: Agentuity Command Runner
Author: agentuity

byagentuity

•

Developer Tools

Executes development commands and parses raw output into structured, actionable error reports with precise file locations.

The Agentuity Command Runner enhances Claude Code by providing a robust framework for executing and analyzing common development tasks like building, testing, linting, and formatting across multiple ecosystems. It automatically detects the appropriate runtime—including Bun, NPM, Go, and Cargo—based on project lockfiles and intelligently parses complex console output into deduplicated, classified error tables. By extracting precise file locations and error types, it allows AI agents to quickly identify, prioritize, and fix regressions or syntax issues without the need for manual log inspection.

Key Features

01Automatic runtime detection for Bun, Node.js, Go, Rust, and Python ecosystems

02Intelligent error classification including type errors, syntax errors, and test failures

0318 GitHub stars

04Precise file and line number extraction from standard tool outputs

05Automatic deduplication and prioritization of critical errors over warnings

06Structured output formatting with status, duration, and exit code summaries

Use Cases

01Performing project-wide type checking and linting with prioritized actionable results

02Standardizing build and installation outputs across monorepos using diverse package managers

03Automating the fix-test-repeat loop by providing structured reports of test failures

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add agentuity/sdk agentuity-command-runner

For use in Claude.ai and ChatGPT

Download Skill