Run golden tests from tests/*.eval.json and generate a JSON pass/fail report.
npx skills add https://github.com/kcleto-ai/growth-os-skills --skill gos-eval
Please help me install this Agent Skill. Skill: gos-eval Source: https://github.com/kcleto-ai/growth-os-skills/tree/main/.claude/skills/gos-eval Command: npx skills add https://github.com/kcleto-ai/growth-os-skills --skill gos-eval
The gos-eval skill provides a deterministic mechanism to run golden tests against Growth OS tooling, capturing failures and producing structured reports to ensure tooling health.
Execute the evaluator with the script .claude/skills/gos-eval/scripts/eval.py to run all fixtures.
High-intent search queries and answers about installing and using this skill.
Deterministic golden tests validate tooling health by executing evaluation scripts against test fixtures, checking exit codes, stdout contents, and JSON status to generate a structured pass/fail report for CI and regression checks.
Validating exit codes and stdout output for CI regression checks involves running evaluation scripts that enforce strict JSON output, capturing pass/fail status per component to generate an aggregated JSON health report.
Test fixtures for automated validation must be strict JSON files placed in the tests/*.eval.json path, enabling the evaluator to execute corresponding Python scripts and verify exit codes, stdout contents, and JSON status reliably.
The golden test evaluator does produce structured JSON reports for automated auditing, aggregating pass/fail outcomes per component based on exit codes, stdout contents, and JSON status validation for reliable tooling health verification.
Strict JSON output is needed for deterministic testing workflows to guarantee reliable tooling health verification, ensuring that exit codes, stdout content checks, and pass/fail status remain consistent and parseable across test runs.
aws-samplesDesign test strategies for AI agents covering unit, integration, and memory tests.
SjdjdiejdrirhdkjejGenerate regression tests from bug reports and detected test frameworks.
thec3ukValidate and enforce global commenting standards across Claude Code projects.
kylesnowschwartzCapture stdout, stderr, and exit codes for CLI golden-file QA.
