Constructor Fabric
Official@constructorfabric
A unified fabric that automates the entire XaaS software development lifecycle from planning to building to running services in prod
Agent Skills by Constructor Fabric
Showing 15 vetted skills indexed across 1 GitHub repositories.
playwright-cli
Drive browsers from the command line to navigate, interact, snapshot, and debug web pages.
file-bug-insight
File Insight product defects as triaged GitHub issues with reproduction evidence.
quality-vector-tests
Write vector-attributed test scenarios tracing PRD requirements to executable test suites.
metric-test
Author and validate YAML fixture and pytest specs for data-path metric tests.
stand-ui-test
Write and review Playwright browser tests against a deployed Insight stand with real Keycloak sign-in.
pr-description
Writes structured pull request bodies for GitHub PRs in the insight repository.
stand-api-test
Write and review HTTP contract tests against a deployed gateway with real Keycloak sessions.
probe-merged-change
Runs exploratory quality testing on merged changes across five quality vectors.
api-test
Redirects legacy analytics HTTP contract test references to the current stand-api-test and metric-test suites.
scope-feature-tests
Plans risk-ordered test scopes for Insight features grounded in code and data flow.
connector
Create, test, validate, and deploy Airbyte-based data connectors for the Insight ingestion platform.
drive-ui
Drives the Insight web UI in a real browser to verify behavior and capture defect evidence.
insight-stand
Operate the Insight compose stand and run the deployed-stand test suite.
stand-scenarios
Converts product scenario documents into testable claims with oracles, layers, and priorities.
check-dbt-conventions
Audits dbt models and connector configurations against ClickHouse ReplacingMergeTree pipeline conventions.
Frequently Asked Questions About Constructor Fabric
FAQPage SchemaWhat tasks can I accomplish with Constructor Fabric's skills?▼
You can write and review Playwright UI journeys and HTTP contract tests for the Insight stand, author metric data-path specs, plan QA coverage from product scenarios, audit dbt models against ClickHouse conventions, create and deploy Insight connectors, and file triaged GitHub defect issues.
Who are these skills designed for?▼
QA engineers, test automation engineers, and data engineers working on the Insight analytics platform. They suit contributors writing committed tests under tests/stand/ and tests/datapath/, reviewers auditing dbt and connector configurations, and maintainers triaging defects or drafting pull request descriptions.
How do the test skills run in practice?▼
Stand tests target a deployed Insight environment brought up with ./dev-compose.sh test-stand up, seeded, then exercised via pytest-based suites with real Keycloak sessions. Interactive browser work uses the playwright-cli binary, while metric specs run against a compose test-stand instance.
What are the prerequisites for using these skills?▼
A running Insight stand (local compose/kind or shared remote), seeded bronze data, and valid persona credentials via Keycloak are required for stand suites. dbt auditing needs the .cf-studio/config/rules/architecture.md conventions, and defect filing targets the public constructorfabric/insight GitHub repository.
Do these skills handle defect reporting and PR documentation?▼
Yes. file-bug-insight drafts, confirms, and creates observation-only GitHub issues with evidence scrubbed of internal detail, while pr-description writes reviewer-friendly PR bodies with labeled Why/What/Verified sections. Both enforce the public-repo rule against production-derived information.