govkit

Coach a Product Manager through creating or improving an Epic via a structured, problem-first interview — core metadata, a rigorous problem statement, impacted personas, business alignment, measurable success metrics, evidence, MVP scope, risks, and NFRs, plus GenAI evaluation criteria when the epic involves model behavior. Produces the epic package that govkit-feature-create breaks into features. Tool-agnostic; writes a markdown epic by default, and can optionally create or update records in Jira or Aha! after explicit confirmation. Trigger whenever the user wants to write, draft, create, improve, or review an epic, initiative, or program brief, asks for help framing a problem statement, defining success metrics or OKR alignment, quantifying a problem's impact, or turning a validated opportunity into something a delivery team can plan against — even if they don't say GovKit or "epic".

Build a visual, scored feature map of a whole epic, release or spec corpus — a chain diagram of how artifacts flow between features, one card per feature with its Gherkin, NFRs and evaluation criteria, a GovKit Development Token badge on every feature showing whether it is ready for AI-assisted coding, and optionally a size badge per feature (scenario complexity distribution) with MVP/V1/V2 slice views from Gherkin tags. Ingests from Jira, Aha!, or a repo directory of feature specs, and merges a tracker record with repo-resident Gherkin. Trigger whenever the user asks to map, visualise, chart, or diagram an epic or set of features, wants a readiness dashboard or portfolio view, asks to score, size, or badge many features at once, wants to see how features connect or what produces and consumes what, asks where a release is weakest or what is blocking delivery, or wants to see how big features are or what the MVP slice costs across a corpus — even if they don't say "feature map" or name GovKit.

Validate whether an approved feature package (Gherkin, NFRs, evaluation criteria) is complete, consistent, repo-aware, and safe for AI-assisted coding. The repo-side readiness gate that runs after the feature refinement. Tool-agnostic;

Review generated feature specs (Gherkin, NFRs, evaluation criteria) before GovKit execution. Use for refinement conversations where Product, QA, and Engineering improve a draft together before AI-assisted coding starts. Tool-agnostic; works with any generator (e.g. Aha!) and any tracker (e.g. Azure DevOps, Jira). Trigger whenever the user mentions Gherkin, acceptance criteria, feature refinement, 3 Amigos, Draft 0, Development Token, spec review, PM pre-review, QA evidence review, or asks to review, score, or rewrite a feature spec before coding begins — even if they don't name GovKit explicitly. Also provides a non-interactive batch scoring mode that emits one JSON verdict per feature; govkit-feature-map calls it to badge a whole corpus.

Emit structured Tier 1 metric events (NDJSON) from a GovKit-governed repository's exhaust — feature packages, .govkit/marker.json, eval_criteria.yaml, CI gate runs, PR exports, and git history. Use this skill whenever the user wants to compute, emit, export, or audit delivery/quality metrics from a GovKit repo: spec completeness scores, gate-readiness audits, metric event streams for an aggregator, velocity/quality pair inputs, or "which features aren't gate-ready." Trigger on mentions of GovKit metrics, metric events, telemetry emission, feature package audit, spec completeness, or AIPOS Tier 1 metrics — even if the user doesn't name this skill.

Generate synthetic test data for a GovKit feature package using Python Faker — a seeded, repeatable generator script plus committed data files derived from the feature's Gherkin scenarios. Use whenever the user asks to "generate test data", "create synthetic data", "seed data", "fixtures", "fake data", "sample records", or "load-test data" for a feature, work item, or acceptance.feature — even if they don't say GovKit or Faker. Also use after a feature package passes readiness and the team needs data to exercise the scenarios. Works for any target stack; only the generator itself is Python.

Author the workflow source behind a customer journey — turn evidence, specs and conversations into a workflow.json naming the outcome, the ordered and branching activities, the actors, the handoffs where work changes hands, and qualified references into the canonical Gherkin that governs each step. Generates L1 journey, L2 collaboration and L3 behavior views from that one file, so a Rule shown at four steps is authored once and no second diagram is maintained by hand. Reports behavior in the corpus that no step touches, and asks only about gaps that would change the journey. Trigger whenever the user asks to map a workflow or customer journey, show how a process flows end to end, work out who does what and where handoffs happen, see which rules apply at which step, check whether a journey covers the behavior that exists, or asks "what does this process actually look like" — even if they don't say GovKit or "workflow".