Name: crux-test
Author: zotoio

Prompt file imported from zotoio/CRUX-Compress (.cursor/commands/crux-test.md). Copyright stays with the author.

crux-test

Run the CRUX command suite via pytest. Test cases live in evals/test_r_crux_command_suite.py; LLM-driven scenarios are marked llm_driven and skipped by default in CI.

Usage

Command	Description
`python3 scripts/run_crux_command_suite.py`	All deterministic tests
`python3 scripts/run_crux_command_suite.py --smoke`	Fast CI smoke subset
`python3 -m pytest evals/test_r_crux_command_suite.py -m crux_command_smoke -v`	Smoke, verbose

Historical Scenario → Pytest Mapping

Historical /crux-test scenario	Pytest class
1 — Compression Test	`TestCompressionRoundtrip`
2 — Decompression Test	`TestDecompressionUnderstanding` (llm_driven)
3 — Token Estimation Test	`TestTokenEstimation`
4 — Checksum Test	`TestChecksumDeterminism`
5 — Install Script Test	`TestInstallScript`
6 — Semantic Validation Test	`TestSemanticValidation` (llm_driven)
7 — Special Characters Test	`TestSpecialCharacters`
8 — Crux-Compress Command Test	`TestCruxCompressWorkflow`
9 — Semantic Stability / Drift Detection	`TestSemanticStabilityDriftDetection`
10 — Force Recompression Test	`TestForceRecompression`

Install crux-test and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "prompt": "# crux-test\n\nRun the CRUX command suite via pytest. Test cases live in\n`evals/test_r_crux_command_suite.py`; LLM-driven scenarios are marked\n`llm_driven` and skipped by default in CI.\n\n## Usage\n\n| Command | Description |\n|---------|-------------|\n| `python3 scripts/run_crux_command_suite.py` | All deterministic tests |\n| `python3 scripts/run_crux_command_suite.py --smoke` | Fast CI smoke subset |\n| `python3 -m pytest evals/test_r_crux_command_suite.py -m crux_command_smoke -v` | Smoke, verbose |\n\n## Historical Scenario → Pytest Mapping\n\n| Historical /crux-test scenario | Pytest class |\n|-------------------------------|--------------|\n| 1 — Compression Test | `TestCompressionRoundtrip` |\n| 2 — Decompression Test | `TestDecompressionUnderstanding` *(llm_driven)* |\n| 3 — Token Estimation Test | `TestTokenEstimation` |\n| 4 — Checksum Test | `TestChecksumDeterminism` |\n| 5 — Install Script Test | `TestInstallScript` |\n| 6 — Semantic Validation Test | `TestSemanticValidation` *(llm_driven)* |\n| 7 — Special Characters Test | `TestSpecialCharacters` |\n| 8 — Crux-Compress Command Test | `TestCruxCompressWorkflow` |\n| 9 — Semantic Stability / Drift Detection | `TestSemanticStabilityDriftDetection` |\n| 10 — Force Recompression Test | `TestForceRecompression` |", "variables": [], "notes": "Run the CRUX command suite via pytest. Test cases live in `evals/test_r_crux_command_suite.py`; LLM-driven scenarios are marked `llm_driven` and skipped by default in CI.", "metadata": { "source": { "provider": "github-cursor-commands", "repository": "https://github.com/zotoio/CRUX-Compress", "path": ".cursor/commands/crux-test.md", "ref": "0ae083e7909b03f37c045083ae6584565a6f18b8", "url": "https://github.com/zotoio/CRUX-Compress/blob/0ae083e7909b03f37c045083ae6584565a6f18b8/.cursor/commands/crux-test.md", "key": "zotoio/CRUX-Compress/.cursor/commands/crux-test.md" } } }

crux-test

About

crux-test

Usage

Historical Scenario → Pytest Mapping

Use it

Install crux-test and it becomes one more candidate the router can pick - when it fits.

Manifest

Reviews