{
  "uid": "cogitave.learn.testing-and-quality.introduction",
  "kind": "moduleUnit",
  "href": "/modules/testing-and-quality/introduction/",
  "title": "Introduction",
  "summary": "",
  "type": null,
  "products": [],
  "roles": [],
  "levels": [],
  "subjects": [],
  "headings": [
    "Why this matters for you",
    "What you will be able to do"
  ],
  "source": "Most teams treat a test as a courtesy - something you add if there is time.\nCogitave does not. Here a test is **release-blocking evidence**: an agent or a\nhuman advances a change from `review` to `done` only when the test gates pass, so\na missing test is a missing piece of the Definition of Done, not a nice-to-have.\n\nTwo standards own this together, and they are separated on purpose:\n\n- The [Testing & Quality Standard](../../../../standards/docs/standards/testing-quality.md)\n  is the **policy** - what is required: the test-shape stance, which test types\n  are mandatory on which surface, the coverage thresholds, the mutation gate, and\n  the flaky-test quarantine rules.\n- The [Test Harness Reference](../../../../standards/docs/standards/test-harness.md)\n  is the **runnable** side - the per-language tool matrix, the exact commands, and\n  the map from each CI gate to its Definition-of-Done item.\n\nAlongside them sits the\n[Code Craftsmanship Standard](../../../../standards/docs/standards/code-craftsmanship.md),\nwhich is the **quality bar review enforces**: SOLID, KISS, YAGNI, and DRY are not\nposters on a wall here - they are review criteria and eval-gate criteria, applied\nto code that humans write and to code that agents generate.\n\n> [!IMPORTANT]\n> Read policy from the standard, wire it up from the harness. Never paraphrase a\n> threshold from memory when the source is one link away - the numbers are the\n> gate.\n\n## Why this matters for you\n\nYou are working inside a control plane, where every change is auditable evidence.\nThe point of this module is not to admire the rules but to let you **test your own\nchange and pass the quality gate**: pick the right test tier, hit the diff-coverage\ngate, keep the suite deterministic, and write code a reviewer will not send back.\n\n## What you will be able to do\n\n- State the test-shape stance and the rule that overrides the ratios.\n- Name what makes coverage a floor, not a goal - and why diff coverage is the gate.\n- Use the harness to run the right tier, and read the gate-to-DoD map.\n- Pass the craftsmanship review bar without applying any principle dogmatically.\n",
  "partOf": "cogitave.learn.testing-and-quality",
  "durationInMinutes": 4,
  "quiz": null
}