skills-driven-development

Close implementation loops with target-bound independent evidence and a portable proof receipt. Use when an agent changes a real system and must earn a completion claim that survives the session.

First fixture skill for skdd add tests. Use when testing drop installation.

Second fixture skill for skdd add tests. Use when testing multi-skill drops.

Safely backfill historical data for a pipeline stage — scope the date range, run a dry-run, execute in chunks to avoid overwhelming the database, and verify row counts after each chunk. Use when historical data is missing, when a pipeline stage is added retroactively, or when asked to "backfill the {stage} table from {start_date}".

Add or extend data validation checks in the pipeline when data quality issues are detected — null-percentage thresholds, value ranges, referential integrity, and row-count sanity checks. Use when a pipeline breaks on bad data, when onboarding a new dataset, or when asked to "add a quality check for {column}".

Onboard a new raw dataset into the data platform — sniff the schema, generate a profiling notebook, create the ingestion job, and add an entry to the data dictionary. Use when adding a new data source, when a partner delivers a new CSV/Parquet drop, or when asked to "onboard the {name} dataset".

Record a model training run — log parameters, metrics, and artifact paths, then append a structured row to the experiments ledger. Use when a training run completes, when asked to "log this experiment", or when comparing runs to decide which model to promote.

Scaffold a new transform stage in the data pipeline — create the transformation script, schema contract, idempotency logic, and tests. Use when adding a new dbt model or pandas transform, when a new business metric needs a dedicated stage, or when asked to "add a {name} stage to the pipeline".

Audit shipctl's public interface before a release — compare flags, exit codes, and output formats against the previous tag to catch accidental breaking changes. Use when preparing a release, when asked to "check for breaking changes", or as a gate before tagging a new version.

Add a new target triple to the shipctl cross-platform build matrix — CI YAML, Makefile, and a smoke test. Use when adding support for a new OS/architecture, when a user requests a new platform binary, or when a target triple is missing from the release artifacts.

Add a CLI flag to shipctl end-to-end — argument parser definition, help text, shell completion scripts, documentation, and tests. Use when adding a new option or flag, when extending an existing subcommand, or when asked to "add a --flag-name option".

Regenerate man pages and shell completion scripts from the shipctl command tree whenever commands, flags, or subcommands change. Use when a new command or flag is added, when the help text changes, or when asked to "update the man pages" or "sync completions".

Cut a versioned release for shipctl — bump the version, generate a changelog from conventional commits, create and push the git tag, and draft a GitHub release with cross-compiled binaries attached. Use when preparing a new release, when asked to "cut a release", or when the version needs bumping after a sprint.

Create or update a reusable agent skill. Use when you notice a repeated pattern, when a workflow should be persisted for future sessions, or when asked to forge/create/scaffold a new skill.

Create or update a reusable agent skill. Use when you notice a repeated pattern, when a workflow should be persisted for future sessions, or when asked to forge/create/scaffold a new skill.

A skill with an invalid name. Use when testing that uppercase and underscores are rejected.

A skill whose name does not match its directory. Use when testing that name/dir mismatches are caught.

A fully-populated skill with SkDD lifecycle metadata. Use when validating the full metadata profile including forged-by, usage-count, and status.

A minimal spec-compliant skill. Use when you need the bare minimum to satisfy the Agent Skills spec.

Deploy a preview branch to the staging environment for review. Use when a feature branch is ready for visual review, when someone asks to "deploy this for review", or before opening a PR.

Scaffold a REST API endpoint with route, controller, validation, and tests. Use when creating new API endpoints, adding resources, or when asked to build a new backend route.

Triage a bug report into an actionable GitHub issue with severity, reproduction steps, and assignment. Use when a bug is reported, when error logs need investigation, or when asked to file/triage an issue.

Create a new React component with co-located test, story, and styles. Use when creating new UI components, when asked to scaffold/generate a component, or when building new pages.