flakiness 0.266.0 → 0.268.0

package/lib/cli/cli.js

@@ -969,7 +969,8 @@ var WireTypes;
       "duration",
       "outcome",
       "duration_trend",
-      "flip_rate"
+      "flip_rate",
+      "fail_rate"
     ],
     timelines: [
       "name",
@@ -1033,7 +1034,7 @@ import path7 from "path";
 // ../package.json
 var package_default = {
   name: "@flakiness/monorepo",
-  version: "0.266.0",
+  version: "0.268.0",
   type: "module",
   private: true,
   scripts: {
@@ -2191,15 +2192,15 @@ var BUNDLED_SKILLS = [
     "files": [
       {
         "path": "SKILL.md",
-        "content": "---\nname: flakiness-investigation\ndescription: >-\n  TRIGGER when: CI fails, tests fail, PR checks fail, user asks about test results, regressions, flakiness, or test health.\n  Always use this INSTEAD OF `gh` CLI for investigating test failures \u2014 it distinguishes regressions (caused by the PR) from pre-existing failures and flakes.\n  DO NOT TRIGGER when: the user is asking about non-test CI issues (e.g. build failures, deployment problems).\n---\n\n# Flakiness Investigation\n\nUse the `flakiness` CLI to query and analyze test data from Flakiness.io.\n\n## Prerequisites\n\n- The `flakiness` CLI must be available in PATH (installed via `npm install -g flakiness` or `npx flakiness`).\n- A project must be specified: `--project <org/project>` or set `FLAKINESS_PROJECT` env var.\n- Authentication: run `flakiness auth login` first, or set `FLAKINESS_ACCESS_TOKEN`.\n\n## Core command\n\n```bash\nflakiness list tests --project <org/project> [--pr <number>] [--branch <name>] [--fql <query>] [--sort <axis>] [--sort-dir asc|desc] [--page <n>] [--page-size <n>] [--env-name <value>] [--env-path <value>] [--env-category <value>] [--env-os <value>] [--env-arch <value>] [--env-metadata <key=value>]\n```\n\n### Scope options\n\n| Flag | Effect |\n|------|--------|\n| *(none)* | Default branch \u2014 tests from the day of its head commit (in the project's timezone) |\n| `--pr <number>` | Pull request \u2014 tests from the merge-commit of the PR branch into the target branch |\n| `--branch <name>` | Named branch \u2014 tests from the day of its head commit (in the project's timezone) |\n\n`--pr` and `--branch` are mutually exclusive.\n\n### Environment filter options\n\nFilter results by environment. All flags are repeatable. Multiple values for the same flag are OR'd; different flags are AND'd.\n\n| Flag | Filters by | Example |\n|------|-----------|---------|\n| `--env-name <value>` | Environment name | `--env-name chromium` |\n| `--env-path <value>` | Config file path | `--env-path playwright.config.ts` |\n| `--env-category <value>` | Test category | `--env-category playwright` |\n| `--env-os <value>` | OS name + version | `--env-os \"Ubuntu 22.04\"` |\n| `--env-arch <value>` | CPU architecture | `--env-arch x86_64` |\n| `--env-metadata <key=value>` | User-supplied metadata | `--env-metadata browser=chromium` |\n\nThe output for each test shows environment info in the same `key=value` format:\n- **Env** line: system-collected data with keys matching the `--env-*` flag suffixes (`category=`, `name=`, `path=`, `os=`, `arch=`)\n- **Env Metadata** line (if any): user-supplied key=value pairs\n\nThis makes it easy to copy a value from the output and use it as a filter flag.\n\n### Status semantics with `--pr`\n\nWhen querying a PR, the status values have specific meaning:\n\n| Status | Meaning |\n|--------|---------|\n| `regressed` | Test was passing on the target branch but fails in this PR \u2014 **caused by the PR** |\n| `failed` | Test also fails on the target branch \u2014 **pre-existing failure, not caused by the PR** |\n| `flaked` | Test failed but passed on retry \u2014 flaky, not a real failure |\n| `passed` | Test passes |\n| `skipped` | Test was skipped |\n\n### Sort axes\n\n- `outcome` \u2014 sort by test outcome severity (default)\n- `flip_rate` \u2014 sort by flip rate (how often a test flips between pass/fail)\n- `duration` \u2014 sort by test duration\n- `duration_trend` \u2014 sort by duration trend\n- `name` \u2014 sort alphabetically\n\n### Common queries\n\n| Goal | FQL + flags |\n|------|-------------|\n| Flaky tests | `--fql 'flip>0%' --sort flip_rate --sort-dir desc` |\n| Very flaky (>50%) | `--fql 'flip>50%' --sort flip_rate --sort-dir desc` |\n| Failed tests | `--fql 's:failed' --sort outcome --sort-dir desc` |\n| Regressions | `--fql 's:regressed'` |\n| All broken tests | `--fql 'status:(failed, regressed)'` |\n| Slow tests | `--fql 'd>5s' --sort duration --sort-dir desc` |\n| Tests matching error | `--fql '$timeout'` |\n| Tests in file | `--fql 'f:login.spec.ts'` |\n| Tests on Linux | `--env-os \"Ubuntu 22.04\"` |\n| Tests for chromium project | `--env-name chromium` |\n| Tests with specific metadata | `--env-metadata browser=firefox` |\n| Linux or macOS failures | `--env-os \"Ubuntu 22.04\" --env-os \"Darwin 24.0\" --fql 's:failed'` |\n\n## FQL (Filter Query Language)\n\nFull reference: [references/fql.md](references/fql.md)\n\nKey rules:\n- Multiple tokens combine with AND: `s:failed f:e2e` means \"failed AND in e2e files\"\n- Prefix with `-` to exclude: `-#smoke` excludes smoke-tagged tests\n- Same filter type uses OR: `status:(failed, regressed)` means \"failed OR regressed\"\n- Quote values with spaces: `f:'tests/e2e checkout'`\n\n### Filter types\n\n| Filter | Syntax | Example |\n|--------|--------|---------|\n| Text search | `<text>` | `login` |\n| Status | `s:<status>` | `s:failed`, `status:(failed, regressed)` |\n| File | `f:<path>` | `f:login.spec.ts` |\n| Error | `$<text>` | `$timeout` |\n| Tag | `#<tag>` | `#smoke` |\n| Duration | `d><time>` | `d>2s`, `d<=500ms` |\n| Flip rate | `flip><pct>` | `flip>0%`, `fr>50%` |\n| Annotation | `@<type>` | `@skip` |\n\n### Status values\n\n`passed`, `failed`, `flaked`, `skipped`, `regressed`\n\n## Workflow: Fix My PR Tests\n\nWhen a user asks to fix failing tests in a PR, follow these steps:\n\n1. **Find the project slug:** Search the codebase for `flakinessProject` to find the `--project` value. It is typically configured in a test reporter config (e.g. `playwright.config.ts`, `jest.config.ts`) as `flakinessProject: 'org/project'`.\n\n2. **Fetch regressions from the PR:**\n   ```bash\n   flakiness list tests --project <org/project> --pr <number> --fql 's:regressed' --page-size 50\n   ```\n   Tests with status `regressed` were passing on the target branch but fail in this PR \u2014 these are **caused by the PR** and must be fixed.\n\n3. **Analyze the output:** Look at the error messages, file paths, and test names to understand what the PR broke.\n\n4. **Fix the regressions** by reading the reported file paths and error messages, then making targeted code changes.\n\n5. **Optionally check pre-existing failures:** Tests with status `failed` also fail on the target branch and are not caused by the PR. You can list them with:\n   ```bash\n   flakiness list tests --project <org/project> --pr <number> --fql 's:failed' --page-size 50\n   ```\n   These are informational \u2014 fixing them is a bonus, not a requirement.\n\n6. **Ignore flakes:** Tests with status `flaked` failed but passed on retry \u2014 they are flaky and not actionable in the context of a PR fix.\n\n## Workflow tips\n\n1. Start broad: `flakiness list tests --project <org/project> --page-size 20`\n2. Filter down with FQL based on what you're investigating\n3. Use `--page-size 50` or higher to see more results at once\n4. Combine filters: `--fql 's:failed $timeout f:e2e -#smoke'`\n\nMore recipes: [references/recipes.md](references/recipes.md)\n"
+        "content": "---\nname: flakiness-investigation\ndescription: >-\n  TRIGGER when: CI fails, tests fail, PR checks fail, user asks about test results, regressions, flakiness, or test health.\n  Always use this INSTEAD OF `gh` CLI for investigating test failures \u2014 it distinguishes regressions (caused by the PR) from pre-existing failures and flakes.\n  DO NOT TRIGGER when: the user is asking about non-test CI issues (e.g. build failures, deployment problems).\n---\n\n# Flakiness Investigation\n\nUse the `flakiness` CLI to query and analyze test data from Flakiness.io.\n\n## Prerequisites\n\n- The `flakiness` CLI must be available in PATH (installed via `npm install -g flakiness` or `npx flakiness`).\n- A project must be specified: `--project <org/project>` or set `FLAKINESS_PROJECT` env var.\n- Authentication: run `flakiness auth login` first, or set `FLAKINESS_ACCESS_TOKEN`.\n\n## Core command\n\n```bash\nflakiness list tests --project <org/project> [--pr <number>] [--branch <name>] [--fql <query>] [--sort <axis>] [--sort-dir asc|desc] [--page <n>] [--page-size <n>] [--env-name <value>] [--env-path <value>] [--env-category <value>] [--env-os <value>] [--env-arch <value>] [--env-metadata <key=value>]\n```\n\n### Scope options\n\n| Flag | Effect |\n|------|--------|\n| *(none)* | Default branch \u2014 tests from the day of its head commit (in the project's timezone) |\n| `--pr <number>` | Pull request \u2014 tests from the merge-commit of the PR branch into the target branch |\n| `--branch <name>` | Named branch \u2014 tests from the day of its head commit (in the project's timezone) |\n\n`--pr` and `--branch` are mutually exclusive.\n\n### Environment filter options\n\nFilter results by environment. All flags are repeatable. Multiple values for the same flag are OR'd; different flags are AND'd.\n\n| Flag | Filters by | Example |\n|------|-----------|---------|\n| `--env-name <value>` | Environment name | `--env-name chromium` |\n| `--env-path <value>` | Config file path | `--env-path playwright.config.ts` |\n| `--env-category <value>` | Test category | `--env-category playwright` |\n| `--env-os <value>` | OS name + version | `--env-os \"Ubuntu 22.04\"` |\n| `--env-arch <value>` | CPU architecture | `--env-arch x86_64` |\n| `--env-metadata <key=value>` | User-supplied metadata | `--env-metadata browser=chromium` |\n\nThe output for each test shows environment info in the same `key=value` format:\n- **Env** line: system-collected data with keys matching the `--env-*` flag suffixes (`category=`, `name=`, `path=`, `os=`, `arch=`)\n- **Env Metadata** line (if any): user-supplied key=value pairs\n\nThis makes it easy to copy a value from the output and use it as a filter flag.\n\n### Status semantics with `--pr`\n\nWhen querying a PR, the status values have specific meaning:\n\n| Status | Meaning |\n|--------|---------|\n| `regressed` | Test was passing on the target branch but fails in this PR \u2014 **caused by the PR** |\n| `failed` | Test also fails on the target branch \u2014 **pre-existing failure, not caused by the PR** |\n| `flaked` | Test failed but passed on retry \u2014 flaky, not a real failure |\n| `passed` | Test passes |\n| `skipped` | Test was skipped |\n\n### Sort axes\n\n- `outcome` \u2014 sort by test outcome severity (default)\n- `flip_rate` \u2014 sort by flip rate (how often a test flips between pass/fail)\n- `fail_rate` \u2014 sort by fail rate (share of commits where the test's terminal status was failing)\n- `duration` \u2014 sort by test duration\n- `duration_trend` \u2014 sort by duration trend\n- `name` \u2014 sort alphabetically\n\n### Common queries\n\n| Goal | FQL + flags |\n|------|-------------|\n| Flaky tests | `--fql 'flip>0%' --sort flip_rate --sort-dir desc` |\n| Very flaky (>50%) | `--fql 'flip>50%' --sort flip_rate --sort-dir desc` |\n| Tests that have failed (terminal failure) | `--fql 'fail>0%' --sort fail_rate --sort-dir desc` |\n| Tests that fail most often (>50%) | `--fql 'fail>50%' --sort fail_rate --sort-dir desc` |\n| Failed tests | `--fql 's:failed' --sort outcome --sort-dir desc` |\n| Regressions | `--fql 's:regressed'` |\n| All broken tests | `--fql 'status:(failed, regressed)'` |\n| Slow tests | `--fql 'd>5s' --sort duration --sort-dir desc` |\n| Tests matching error | `--fql '$timeout'` |\n| Tests in file | `--fql 'f:login.spec.ts'` |\n| Tests on Linux | `--env-os \"Ubuntu 22.04\"` |\n| Tests for chromium project | `--env-name chromium` |\n| Tests with specific metadata | `--env-metadata browser=firefox` |\n| Linux or macOS failures | `--env-os \"Ubuntu 22.04\" --env-os \"Darwin 24.0\" --fql 's:failed'` |\n\n## FQL (Filter Query Language)\n\nFull reference: [references/fql.md](references/fql.md)\n\nKey rules:\n- Multiple tokens combine with AND: `s:failed f:e2e` means \"failed AND in e2e files\"\n- Prefix with `-` to exclude: `-#smoke` excludes smoke-tagged tests\n- Same filter type uses OR: `status:(failed, regressed)` means \"failed OR regressed\"\n- Quote values with spaces: `f:'tests/e2e checkout'`\n\n### Filter types\n\n| Filter | Syntax | Example |\n|--------|--------|---------|\n| Text search | `<text>` | `login` |\n| Status | `s:<status>` | `s:failed`, `status:(failed, regressed)` |\n| File | `f:<path>` | `f:login.spec.ts` |\n| Error | `$<text>` | `$timeout` |\n| Tag | `#<tag>` | `#smoke` |\n| Duration | `d><time>` | `d>2s`, `d<=500ms` |\n| Flip rate | `flip><pct>` | `flip>0%`, `fr>50%` |\n| Annotation | `@<type>` | `@skip` |\n\n### Status values\n\n`passed`, `failed`, `flaked`, `skipped`, `regressed`\n\n## Workflow: Fix My PR Tests\n\nWhen a user asks to fix failing tests in a PR, follow these steps:\n\n1. **Find the project slug:** Search the codebase for `flakinessProject` to find the `--project` value. It is typically configured in a test reporter config (e.g. `playwright.config.ts`, `jest.config.ts`) as `flakinessProject: 'org/project'`.\n\n2. **Fetch regressions from the PR:**\n   ```bash\n   flakiness list tests --project <org/project> --pr <number> --fql 's:regressed' --page-size 50\n   ```\n   Tests with status `regressed` were passing on the target branch but fail in this PR \u2014 these are **caused by the PR** and must be fixed.\n\n3. **Analyze the output:** Look at the error messages, file paths, and test names to understand what the PR broke.\n\n4. **Fix the regressions** by reading the reported file paths and error messages, then making targeted code changes.\n\n5. **Optionally check pre-existing failures:** Tests with status `failed` also fail on the target branch and are not caused by the PR. You can list them with:\n   ```bash\n   flakiness list tests --project <org/project> --pr <number> --fql 's:failed' --page-size 50\n   ```\n   These are informational \u2014 fixing them is a bonus, not a requirement.\n\n6. **Ignore flakes:** Tests with status `flaked` failed but passed on retry \u2014 they are flaky and not actionable in the context of a PR fix.\n\n## Workflow tips\n\n1. Start broad: `flakiness list tests --project <org/project> --page-size 20`\n2. Filter down with FQL based on what you're investigating\n3. Use `--page-size 50` or higher to see more results at once\n4. Combine filters: `--fql 's:failed $timeout f:e2e -#smoke'`\n\nMore recipes: [references/recipes.md](references/recipes.md)\n"
       },
       {
         "path": "references/fql.md",
-        "content": "# FQL Reference\n\nUse `--fql <query>` with `flakiness list tests` to filter tests before sorting and pagination.\n\n## Core rules\n\n- Multiple tokens are combined with `AND` by default.\n- Prefix a token with `-` to exclude matches.\n- Use quotes for spaces or special characters.\n- Repeating the same filter type behaves like `OR`, for example `status:(failed, regressed)`.\n\n## Filter types\n\n| Filter | Examples | Meaning |\n| --- | --- | --- |\n| Test text | `login` | Matches test titles, suite titles, and file path |\n| Status | `s:failed`, `status:(failed, regressed)` | Matches test status |\n| File | `f:login.spec.ts`, `file:e2e` | Matches file path |\n| Line | `:120`, `>200`, `<=50` | Matches line number |\n| Error | `$timeout`, `$'network error'` | Matches error text |\n| Annotation | `@skip`, `@'fails in ci'` | Matches annotation type or description |\n| Tag | `#smoke` | Matches test tags |\n| Duration | `d>2s`, `duration<=500ms` | Matches duration |\n| Flip rate | `flip>0%`, `fr<=50%` | Matches flip rate |\n\n## Status values\n\n- `passed`\n- `failed`\n- `flaked`\n- `skipped`\n- `regressed`\n\n## Examples\n\n```fql\ns:flaked\nstatus:(failed, regressed)\nf:e2e d>5s\nflip>0%\nfr>50%\n$timeout -#smoke\n@skip f:'tests/e2e checkout'\n```\n\n## Quoting\n\n- Single and double quotes both work.\n- Escape quotes and backslashes with JSON-style escaping.\n\n```fql\nf:'some path'\n$\"say \\\"hello\\\"\"\n'can\\'t touch this'\n```\n\n## Flip rate\n\nValues are percentages (0\u2013100). The `%` suffix is optional.\n\n```fql\nflip>0%\nflip>=50\nfr<=25%\n```\n\n## Duration units\n\n- `ns`\n- `ms` or no suffix\n- `s`\n- `m`\n- `h`\n\n```fql\nd>100\nd>1.5s\nduration<=2m\n```\n"
+        "content": "# FQL Reference\n\nUse `--fql <query>` with `flakiness list tests` to filter tests before sorting and pagination.\n\n## Core rules\n\n- Multiple tokens are combined with `AND` by default.\n- Prefix a token with `-` to exclude matches.\n- Use quotes for spaces or special characters.\n- Repeating the same filter type behaves like `OR`, for example `status:(failed, regressed)`.\n\n## Filter types\n\n| Filter | Examples | Meaning |\n| --- | --- | --- |\n| Test text | `login` | Matches test titles, suite titles, and file path |\n| Status | `s:failed`, `status:(failed, regressed)` | Matches test status |\n| File | `f:login.spec.ts`, `file:e2e` | Matches file path |\n| Line | `:120`, `>200`, `<=50` | Matches line number |\n| Error | `$timeout`, `$'network error'` | Matches error text |\n| Annotation | `@skip`, `@'fails in ci'` | Matches annotation type or description |\n| Tag | `#smoke` | Matches test tags |\n| Duration | `d>2s`, `duration<=500ms` | Matches duration |\n| Flip rate | `flip>0%`, `fr<=50%` | Matches flip rate |\n| Fail rate | `fail>0%`, `fail<=50%` | Matches fail rate |\n\n## Status values\n\n- `passed`\n- `failed`\n- `flaked`\n- `skipped`\n- `regressed`\n\n## Examples\n\n```fql\ns:flaked\nstatus:(failed, regressed)\nf:e2e d>5s\nflip>0%\nfr>50%\nfail>0%\n$timeout -#smoke\n@skip f:'tests/e2e checkout'\n```\n\n## Quoting\n\n- Single and double quotes both work.\n- Escape quotes and backslashes with JSON-style escaping.\n\n```fql\nf:'some path'\n$\"say \\\"hello\\\"\"\n'can\\'t touch this'\n```\n\n## Flip rate\n\nValues are percentages (0\u2013100). The `%` suffix is optional.\n\n```fql\nflip>0%\nflip>=50\nfr<=25%\n```\n\n## Fail rate\n\nFraction of commits where the test's terminal status was failing (`unexpected` or `regressed`). Flakes do not count as failures because the test eventually passed. Skipped commits do not count toward the denominator. Values are percentages (0\u2013100); the `%` suffix is optional.\n\n```fql\nfail>0%\nfail>=50\nfail<=25%\n```\n\n## Duration units\n\n- `ns`\n- `ms` or no suffix\n- `s`\n- `m`\n- `h`\n\n```fql\nd>100\nd>1.5s\nduration<=2m\n```\n"
       },
       {
         "path": "references/recipes.md",
-        "content": "# Investigation Recipes\n\nReplace `myorg/myproject` with the target project slug, or set `FLAKINESS_PROJECT`.\n\n## Query a pull request\n\nShows tests from the merge-commit of the PR branch into the target branch.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42\n```\n\n## Find PR regressions (caused by the PR)\n\nTests marked `regressed` were passing on the target branch but fail in this PR.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42 --fql 's:regressed'\n```\n\n## Find pre-existing failures in a PR\n\nTests marked `failed` also fail on the target branch \u2014 not caused by the PR.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42 --fql 's:failed'\n```\n\n## Find all broken tests in a PR\n\nIncludes both PR-caused regressions and pre-existing failures.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42 --fql 'status:(failed, regressed)' --page-size 50\n```\n\n## Query a specific branch\n\nShows tests from the day of the branch's head commit, in the project's timezone.\n\n```bash\nflakiness list tests --project myorg/myproject --branch feature/login\n```\n\n## Find all flaky tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'flip>0%' --sort flip_rate --sort-dir desc\n```\n\n## Find the most flaky tests (>50% flip rate)\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'flip>50%' --sort flip_rate --sort-dir desc\n```\n\n## Show failed tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:failed' --sort outcome --sort-dir desc\n```\n\n## Show regressions\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:regressed'\n```\n\n## Show all currently broken tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'status:(failed, regressed)' --sort outcome --sort-dir desc\n```\n\n## Show flaked tests (passed on retry)\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:flaked' --sort flip_rate --sort-dir desc\n```\n\n## Find slow tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'd>5s' --sort duration --sort-dir desc\n```\n\n## Find tests by error text\n\n```bash\nflakiness list tests --project myorg/myproject --fql '$timeout'\nflakiness list tests --project myorg/myproject --fql '$\"network error\"'\n```\n\n## Combine filters\n\nFailed tests in e2e files, excluding smoke-tagged tests:\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:failed f:e2e -#smoke'\n```\n\n## Narrow to a specific file\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'f:login.spec.ts'\n```\n\n## Find tests with a specific annotation\n\n```bash\nflakiness list tests --project myorg/myproject --fql '@skip'\nflakiness list tests --project myorg/myproject --fql '@fixme'\n```\n\n## Filter by environment\n\n### Show tests for a specific OS\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\"\n```\n\n### Show tests for a specific environment name\n\n```bash\nflakiness list tests --project myorg/myproject --env-name chromium\n```\n\n### Show tests for a specific architecture\n\n```bash\nflakiness list tests --project myorg/myproject --env-arch arm64\n```\n\n### Show tests for multiple OSes (OR)\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\" --env-os \"Darwin 24.0\"\n```\n\n### Combine env filter with FQL\n\nFailed tests on Linux only:\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\" --fql 's:failed'\n```\n\n### Filter by user-supplied metadata\n\n```bash\nflakiness list tests --project myorg/myproject --env-metadata browser=firefox\n```\n\n### Combine multiple env filters (AND)\n\nTests on Linux with arm64 architecture:\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\" --env-arch arm64\n```\n"
+        "content": "# Investigation Recipes\n\nReplace `myorg/myproject` with the target project slug, or set `FLAKINESS_PROJECT`.\n\n## Query a pull request\n\nShows tests from the merge-commit of the PR branch into the target branch.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42\n```\n\n## Find PR regressions (caused by the PR)\n\nTests marked `regressed` were passing on the target branch but fail in this PR.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42 --fql 's:regressed'\n```\n\n## Find pre-existing failures in a PR\n\nTests marked `failed` also fail on the target branch \u2014 not caused by the PR.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42 --fql 's:failed'\n```\n\n## Find all broken tests in a PR\n\nIncludes both PR-caused regressions and pre-existing failures.\n\n```bash\nflakiness list tests --project myorg/myproject --pr 42 --fql 'status:(failed, regressed)' --page-size 50\n```\n\n## Query a specific branch\n\nShows tests from the day of the branch's head commit, in the project's timezone.\n\n```bash\nflakiness list tests --project myorg/myproject --branch feature/login\n```\n\n## Find all flaky tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'flip>0%' --sort flip_rate --sort-dir desc\n```\n\n## Find the most flaky tests (>50% flip rate)\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'flip>50%' --sort flip_rate --sort-dir desc\n```\n\n## Find tests that have ever failed (terminal failure, retries exhausted)\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'fail>0%' --sort fail_rate --sort-dir desc\n```\n\n## Find tests that fail more than half the time\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'fail>50%' --sort fail_rate --sort-dir desc\n```\n\n## Show failed tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:failed' --sort outcome --sort-dir desc\n```\n\n## Show regressions\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:regressed'\n```\n\n## Show all currently broken tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'status:(failed, regressed)' --sort outcome --sort-dir desc\n```\n\n## Show flaked tests (passed on retry)\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:flaked' --sort flip_rate --sort-dir desc\n```\n\n## Find slow tests\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'd>5s' --sort duration --sort-dir desc\n```\n\n## Find tests by error text\n\n```bash\nflakiness list tests --project myorg/myproject --fql '$timeout'\nflakiness list tests --project myorg/myproject --fql '$\"network error\"'\n```\n\n## Combine filters\n\nFailed tests in e2e files, excluding smoke-tagged tests:\n\n```bash\nflakiness list tests --project myorg/myproject --fql 's:failed f:e2e -#smoke'\n```\n\n## Narrow to a specific file\n\n```bash\nflakiness list tests --project myorg/myproject --fql 'f:login.spec.ts'\n```\n\n## Find tests with a specific annotation\n\n```bash\nflakiness list tests --project myorg/myproject --fql '@skip'\nflakiness list tests --project myorg/myproject --fql '@fixme'\n```\n\n## Filter by environment\n\n### Show tests for a specific OS\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\"\n```\n\n### Show tests for a specific environment name\n\n```bash\nflakiness list tests --project myorg/myproject --env-name chromium\n```\n\n### Show tests for a specific architecture\n\n```bash\nflakiness list tests --project myorg/myproject --env-arch arm64\n```\n\n### Show tests for multiple OSes (OR)\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\" --env-os \"Darwin 24.0\"\n```\n\n### Combine env filter with FQL\n\nFailed tests on Linux only:\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\" --fql 's:failed'\n```\n\n### Filter by user-supplied metadata\n\n```bash\nflakiness list tests --project myorg/myproject --env-metadata browser=firefox\n```\n\n### Combine multiple env filters (AND)\n\nTests on Linux with arm64 architecture:\n\n```bash\nflakiness list tests --project myorg/myproject --env-os \"Ubuntu 22.04\" --env-arch arm64\n```\n"
       }
     ]
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "flakiness",
-  "version": "0.266.0",
+  "version": "0.268.0",
   "private": false,
   "bin": {
     "flakiness": "./lib/cli/cli.js"
@@ -23,8 +23,8 @@
     "@types/debug": "^4.1.13",
     "@types/express": "^4.17.25",
     "gray-matter": "^4.0.3",
-    "@flakiness/server": "0.266.0",
-    "@flakiness/shared": "0.266.0"
+    "@flakiness/server": "0.268.0",
+    "@flakiness/shared": "0.268.0"
   },
   "dependencies": {
     "@flakiness/flakiness-report": "^0.34.0",