# Test Package > Run grok test for Datagrok packages and report results Use when this capability is needed. - Skill: `tomevault-io/test-package` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/test-package` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/test-package/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Docs & Writing - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/test-package --- # Test Package Run `grok test` for one or more Datagrok packages and provide a summary of results. ## Instructions 1. Parse the user's input to extract: - One or more package names (required). These are directory names under `public/packages/`. - `--host` value (default: `localhost`) - Use `--csv` flag to get csv output - Check if npm install is needed - Check if package needs to be rebuilt. Compare /dist/*.js timestamp with the most recent timestamp in *.ts or *.js file in the package. If dist folder is newer, use --skip-build - Any extra `grok test` flags (`--category`, `--test`, `--verbose`, `--skip-build`, `--gui`, `--no-retry`, `--benchmark`, `--tag`) 2. For each package, run tests from the package directory using the Bash tool: ``` cd /c/Repos/Datagrok/public/packages/ && node grok test --host --no-retry [extra flags] ``` Use a timeout of 600000ms (10 minutes). Print output to console. 3. After all packages finish, provide a **summary table** with: - Package name - Total passed / failed / skipped counts (parse csv output file test-results.csv) - List of any failed categories and test names - Overall verdict: PASSED or FAILED 4. Cleanup csv output file --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/datagrok-ai) — claim your Tome and manage your conversions.