How do I install the Labeling Pipelines skill?

Run npx skillmds@latest add scylladb/labeling-pipelines in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Labeling Pipelines safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Labeling Pipelines?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Labeling Pipelines free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Labeling Pipelines?

scylladb (@scylladb) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Labeling Pipelines by scylladb — Agent Skill

Name: Labeling Pipelines
Author: scylladb

Labeling Pipelines

Guides labeling and linting test-case YAML files in SCT with test_metadata sections. Use when bulk-adding test_metadata to multiple test cases, running lint-test-docs to find coverage gaps, or auditing label accuracy across a directory. Triggers: "label pipelines", "add metadata to test cases", "bulk document tests", "find missing test_metadata", "lint test docs".

scylladb c96e8ec 2 files · 5.2 KB Updated Sep 17, 2026

File contents

Labeling Pipelines for SCT

Bulk-add or audit test_metadata: labels across test-case YAML files.

Workflow: Add test_metadata to a test case

Read the test-case YAML to understand what it does
Check test_duration (minutes) → pick duration_class (short <6h, medium 6-24h, long >24h)
Check stress_cmd* fields → add tools to stress_tools
Check server_encrypt, n_db_nodes → add to features
Check cluster_backend → add to supported_backends (or leave null for all)
Set team_ownership to the QA team responsible for maintaining this test (see taxonomy)
Write a 2-4 sentence description explaining what the test validates
Run the lint-test-docs CLI with --test-case-file flag pointing to your file to verify

Note: nemesis_labels appears in older examples but is not currently a validated TestMetadata field (no taxonomy/pydantic support) — omit it from new test_metadata: blocks rather than add unenforced data.

Complete Example

test_metadata:
  description: >-
    Basic longevity test running cassandra-stress write workload at QUORUM
    consistency for ~4 hours on a 6-node single-DC cluster with SisyphusMonkey
    nemesis. Validates cluster stability under moderate write load with chaos.
  test_type: longevity
  tier: tier1
  duration_class: short
  supported_backends:
    - aws
    - gce
    - azure
  stress_tools:
    - cassandra-stress
  workload: write
  features: []
  team_ownership: core-test-infra

Workflow: Find coverage gaps

Run the lint-test-docs CLI command with --missing-only flag to list all test-case YAMLs that have no test_metadata: section.

Workflow: Lint all test cases

Run lint-test-docs with no flags. Exit code 1 if any errors (missing required fields, invalid values). Warnings (cross-reference mismatches) do not fail the run.

Auto-fix

Run lint-test-docs with --fix flag. Auto-corrects: duration_class from test_duration, stress_tools from stress_cmd* fields, multi-dc feature from n_db_nodes, tls-ssl from server_encrypt.

Tier Selection Guide

Tier	When to use
`sanity`	Runs on every commit, <1h, basic smoke
`tier1`	Weekly core regression, 3-24h
`tier2`	Extended regression, run less frequently
`release`	Mandatory for release qualification
`ondemand`	Investigation, niche, or expensive tests

Valid Field Values

See references/taxonomy-values.md for the full list of valid values per field. The authoritative source is always docs/pipeline-labels/taxonomy.yaml.

Validation Rules Quick Reference

Rule	Check	Severity
TD-001	test_metadata section exists	error
TD-002	description >20 chars	error
TD-003	all values valid per pydantic	error
TD-004	test_type matches directory	warning
TD-005	nemesis_labels includes nemesis_class_name	warning
TD-006	stress_tools matches stress_cmd*	warning
TD-007	supported_backends includes cluster_backend	warning
TD-008	duration_class matches test_duration	warning
TD-009	features consistent with config	warning
TD-010	tier appropriate for duration_class	warning
TD-011	supported_backends includes pipeline backend	warning

---name: labeling-pipelinesdescription: Guides labeling and linting test-case YAML files in SCT with test_metadata sections. Use when bulk-adding test_metadata to multiple test cases, running lint-test-docs to find coverage gaps, or auditing label accuracy across a directory. Triggers: "label pipelines", "add metadata to test cases", "bulk document tests", "find missing test_metadata", "lint test docs".---# Labeling Pipelines for SCTBulk-add or audit `test_metadata:` labels across test-case YAML files.## Workflow: Add test_metadata to a test case1. Read the test-case YAML to understand what it does2. Check `test_duration` (minutes) → pick `duration_class` (short <6h, medium 6-24h, long >24h)3. Check `stress_cmd*` fields → add tools to `stress_tools`4. Check `server_encrypt`, `n_db_nodes` → add to `features`5. Check `cluster_backend` → add to `supported_backends` (or leave null for all)6. Set `team_ownership` to the QA team responsible for maintaining this test (see taxonomy)7. Write a 2-4 sentence `description` explaining what the test validates8. Run the `lint-test-docs` CLI with `--test-case-file` flag pointing to your file to verifyNote: `nemesis_labels` appears in older examples but is not currently a validated`TestMetadata` field (no taxonomy/pydantic support) — omit it from new`test_metadata:` blocks rather than add unenforced data.## Complete Example```yamltest_metadata: description: >- Basic longevity test running cassandra-stress write workload at QUORUM consistency for ~4 hours on a 6-node single-DC cluster with SisyphusMonkey nemesis. Validates cluster stability under moderate write load with chaos. test_type: longevity tier: tier1 duration_class: short supported_backends: - aws - gce - azure stress_tools: - cassandra-stress workload: write features: [] team_ownership: core-test-infra```## Workflow: Find coverage gapsRun the `lint-test-docs` CLI command with `--missing-only` flag to list alltest-case YAMLs that have no `test_metadata:` section.## Workflow: Lint all test casesRun `lint-test-docs` with no flags. Exit code 1 if any errors (missing requiredfields, invalid values). Warnings (cross-reference mismatches) do not fail the run.## Auto-fixRun `lint-test-docs` with `--fix` flag. Auto-corrects: `duration_class` from`test_duration`, `stress_tools` from `stress_cmd*` fields, `multi-dc` featurefrom `n_db_nodes`, `tls-ssl` from `server_encrypt`.## Tier Selection Guide| Tier | When to use ||------|-------------|| `sanity` | Runs on every commit, <1h, basic smoke || `tier1` | Weekly core regression, 3-24h || `tier2` | Extended regression, run less frequently || `release` | Mandatory for release qualification || `ondemand` | Investigation, niche, or expensive tests |## Valid Field ValuesSee `references/taxonomy-values.md` for the full list of valid values per field.The authoritative source is always `docs/pipeline-labels/taxonomy.yaml`.## Validation Rules Quick Reference| Rule | Check | Severity ||------|-------|----------|| TD-001 | test_metadata section exists | error || TD-002 | description >20 chars | error || TD-003 | all values valid per pydantic | error || TD-004 | test_type matches directory | warning || TD-005 | nemesis_labels includes nemesis_class_name | warning || TD-006 | stress_tools matches stress_cmd* | warning || TD-007 | supported_backends includes cluster_backend | warning || TD-008 | duration_class matches test_duration | warning || TD-009 | features consistent with config | warning || TD-010 | tier appropriate for duration_class | warning || TD-011 | supported_backends includes pipeline backend | warning |

Frequently asked questions