agentleFS
my-chiefmindmy-chiefmind/ai-maestro/skills/pipeline-quality/SKILL.md

pipeline-quality

Skill6 starsChanged 50 days agoMITView on GitHub

Validate data-pipeline changes against quality contracts: run the pipeline test suite, verify stage idempotency, check extraction fixtures against expected output, confirm schema fixture coverage, and gate migration safety. Use before merging any pipeline change.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run the pipeline area's test command from context.md (the same command the release gate uses) from the project root.

    line 17
  2. Distinctive:

    Run the stage twice on the same input.

    line 28
  3. Distinctive:

    Run the extractor against the fixture.

    line 36
  4. Distinctive:

    If the change includes a database migration:

    line 49
  5. Distinctive:

    Run the migration on a throwaway DB: assert upgrade succeeds, run downgrade, assert state is restored.

    line 53
  6. Very specific:

    Read the project's context.md for pipeline-specific quality rules — QA/confidence thresholds, fixture conventions, entity-specific …

    line 58

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Pipeline Quality

Validates a data pipeline against quality contracts: correctness, idempotency, fixture coverage, and extraction fidelity. Domain-specific rules (confidence thresholds, fixture conventions, per-entity checks) belong in the project's context.md — this skill carries the contract every pipeline shares.

Common contract

1. Test suite

Run the pipeline area's test command from context.md (the same command the release gate uses) from the project root.

  • Must exit 0.
  • Report any failures with test name and failure message.
  • Coverage must meet the threshold the project declares in context.md, if any.

2. Idempotency

For every pipeline stage the change touches:

  • Run the stage twice on the same input.
  • Assert the second run writes nothing — no new DB records, no file changes, no errors.
  • A stage that is not idempotent is a blocking defect.

3. Fixture regression

For every extraction fixture in the test suite:

  • Run the extractor against the fixture.
  • Assert the output matches the expected fixture output exactly.
  • A changed fixture output that was not explicitly updated is a blocking defect.

4. Schema completeness

For every field in the extraction schema:

  • A test fixture must exercise that field with a valid, a null/missing, and an invalid value.
  • Fields without fixture coverage are flagged as warnings.

5. Migration safety

If the change includes a database migration:

  • The downgrade path must be implemented.
  • A NOT NULL column on a non-empty table requires a server default.
  • Run the migration on a throwaway DB: assert upgrade succeeds, run downgrade, assert state is restored.

Domain rules

Read the project's context.md for pipeline-specific quality rules — QA/confidence thresholds, fixture conventions, entity-specific invariants — and apply them alongside the common contract. A project that only needs to add rules should append them in custom/skills/pipeline-quality/OVERLAY.md, which keeps this skill updating underneath; a project that needs a fundamentally different contract can replace it outright with custom/skills/pipeline-quality/SKILL.md.

Output

=== Pipeline Quality ===

Test suite:      PASS / FAIL   (<n> tests, <m> failures)
Idempotency:     PASS / FAIL   (<stages checked>)
Fixture parity:  PASS / FAIL   (<n> fixtures, <m> mismatches)
Schema coverage: PASS / WARN   (<n> fields, <m> without coverage)
Migration:       PASS / FAIL / SKIP

BLOCKERS:
  - <description>

WARNINGS:
  - <description>

VERDICT: SHIP / BLOCK

A BLOCK sends the ticket back to build with the specific failure — same contract as the release-gate skill.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.