Kemetra

speckit-analyze

Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.

speckit-checklist

Generate a custom checklist for the current feature based on user requirements.

speckit-clarify

Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.

speckit-constitution

Create or update the project constitution from interactive or provided principle inputs, ensuring all dependent templates stay in sync.

speckit-implement

Execute the implementation plan by processing and executing all tasks defined in tasks.md

speckit-plan

Execute the implementation planning workflow using the plan template to generate design artifacts.

DevOps & Infrastructure Listed

speckit-specify

Create or update the feature specification from a natural language feature description.

speckit-tasks

Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.

speckit-git-feature

Create a feature branch with sequential or timestamp numbering

speckit-git-commit

Auto-commit changes after a Spec Kit command completes

speckit-git-validate

Validate current branch follows feature branch naming conventions

Package a raised judgment call into a reviewable decision package, then record a named human's answer into the committed readiness artifacts. Use for sign-off packaging or recording a decision. It only transcribes a decision a human supplied.

approval-evidence-pack

Assemble a reviewable evidence pack behind a readiness approval in the Seshat BI repo. Use when someone asks to package the evidence for a stage sign-off. Read-only: it collects committed evidence and never grants the approval.

business-knowledge-interview

After a database is discovered and profiled, interview the business owner to capture the meaning-critical decisions the rest of the Database-to-PBIP flow depends on -- KPI meaning, PII handling, table grain, primary keys, relationships, missing-value rules, and ambiguous financial/quantity/date columns -- and record every outcome in the project Decision Store. Use when someone asks to run the interview, capture business knowledge, or resolve the business questions behind a model in the Seshat BI repo. It asks obvious low-risk items as one batch and critical items individually with explicit named-human approval; it uses masked samples and never shows raw suspected PII by default; it records decisions and STOPS. It does NOT define metric contracts or DAX, does NOT self-grant any approval, and does NOT emit a confidence score.

capabilities

Show the read-only capability inventory for this kit -- each capability by lifecycle state, authority, requirements and provenance. Use when someone asks what the kit can do or what works without a database. Writes nothing and emits no score.

Data & Documents Listed

consumer-data-dictionary

Generate a consumer-facing data dictionary from committed gold-layer artifacts and approved metric contracts. Use when someone asks for a data dictionary or column glossary for report users. Authors documentation only and invents no business meaning.

cross-table-lineage

Trace column- and table-level lineage across bronze/silver/gold in the Seshat BI repo and report the downstream impact of a change. Read-only: it reads committed SQL and mappings, opens no database, and edits nothing.

dagster-orchestration-adapter

Run the medallion sequence UNATTENDED / in CI as a Dagster asset graph -- the scheduler sibling of the retail-orchestrate conductor -- running ONLY already-approved steps behind every gate, and recording what each asset DID as DERIVED run-evidence. Use when someone asks to schedule the pipeline, run the medallion in CI, "orchestrate unattended", or wire Dagster in the Seshat BI repo. Dagster RUNS approved steps; Tower BI (the gate exit code + the named human) decides whether a stage passed. This skill READS committed approvals as the GO signal, EXECUTES approved steps behind STOP / HUMAN-SEAM edges, and WRITES run-evidence; it never defines meaning, never moves a stage to pass, never publishes Power BI (it only TRIGGERS F016), and HARD-STOPS at every human judgment call.

dashboard-design

Design a Power BI dashboard FROM approved metric contracts in the Seshat BI repo. Use when someone asks to design a report/page, bind visuals to metrics, or produce a dashboard layout for a subject area whose semantic model is ready. This skill is HARD-GATED on semantic_model_ready: pass (no approved metric contracts -> no design); it authors reviewable design guidance only -- a layout plan, a visual list, and a visual->contract binding map where every visual binds to exactly one approved contract. It NEVER invents a metric, NEVER publishes, NEVER opens Power BI Desktop or a DB connection, and NEVER calls the Power BI execution adapter (official Power BI MCP / connection; `pbi-cli` no longer preferred) -- that is feature F016. It authors, runs static seshat check, records dashboard_ready at most as warning, and STOPS.

dashboard-intelligence

Coordinate the SHIPPED Seshat dashboard capabilities into one governed journey (spec 123, US2) -- from an APPROVED Report Intent through gap/duplication checks, contract-bound design, report composition, and dashboard QA, STOPPING at the human blueprint review. Use when someone asks to "build/design the dashboard end to end", "coordinate the dashboard work", or drive a subject area from an approved intent to a reviewable design in the Seshat BI repo. The AGENT is the runtime: this skill inspects committed state, picks ONE next allowed action, invokes the existing capability, and re-evaluates. It reuses `retail dashboard-gaps`, `retail dashboard-planner`, `dashboard-design`, the blueprint/visual-spec/ composition templates, visual-to-contract binding, and dashboard QA -- it adds NO new CLI family, NEVER bypasses `semantic_model_ready: pass`, and NEVER self-grants `dashboard_ready: pass`. It fails closed with a named blocker on the first unmet precondition.

dbt-transformation-adapter

Run dbt as the build ENGINE for silver/gold ONLY behind the Mapping Ready gate, and record its run/test/parity output as DERIVED evidence -- never as an approval. Use when someone asks to build silver/gold for a mapped table with dbt, run dbt tests, or check dbt parity against the existing gold tables in the Seshat BI repo. dbt is the engine; Seshat BI is the brain. This skill READS the approved map, EXECUTES approved dbt steps behind the gate, and WRITES evidence; it never defines meaning, never moves a stage to pass, and HARD-STOPS at every human judgment call.

evidence-pack-generator

Compose a single readable 10-section evidence pack for a table/report at the late readiness stages (Semantic Model Ready -> Dashboard Ready -> Publish Ready) in the Seshat BI repo. Use when someone asks to "assemble the evidence pack", "gather everything for the data-owner sign-off", or "show the full late-stage evidence for this table". This is a Product Module, artifact-writing: it READS the already-committed upstream artifacts, renders the ordered 10-section pack index + one-page summary, records each section's status (one of four) with evidence and blockers, and SURFACES the recorded publish_ready state read from readiness-status.yaml -- then STOPS. It invents NO section content (a missing or unfilled source is a blocker, never fabricated), embeds the shipped F013 handoff pack as section 08 without re-authoring it, writes NO approval, moves NO stage to pass, edits NO source artifact, runs NO live DB/PBIP read, and emits NO numeric confidence/health score.

first-hour-compass

Show a NEW-table author the stateful "you are here / next artifact / authoring skill" orientation card for ONE table in the Seshat BI repo. Use when someone asks "where is this table in the readiness journey", "what do I do next for <table>", or "orient me on a new table". READ-ONLY and invoke-and-present only: it parses that table's mappings/<table>/readiness-status.yaml and RENDERS a single-table orientation card (current stage, the next non-pass stage + the artifact it needs + the authoring skill that produces it, plus recorded STOP rows). It RENDERS recorded state (never recomputes a status, advances a stage, writes a pass, infers/back-fills an approval, or fabricates evidence), runs NO validator, opens NO DB connection, and emits NO numeric health/confidence/percent-ready/maturity score (hard rule #9). It is the STATEFUL single-table sibling of the F026 readiness-viewer (multi-table matrix) and the stateful form of the F006 static onboarding checklist.

friendly-pr-reviewer

Render a PLAIN-LANGUAGE narrative summary of one PR's already-produced governance review -- what changed, which readiness stages are pass/warning/blocked (verbatim), which findings are NEW/RESOLVED/pre-existing (when a base identity set is supplied), who must approve, and exactly one next action. Use when someone asks "explain this PR in plain language", "what does this review envelope actually mean", or wants a non-technical summary of a PR's governance findings WITHOUT opening JSON/YAML. It is a NARRATIVE over already-shipped truth (`seshat check --format review`, readiness truth), NOT a merge-safety verdict -- see the boundary note against F025 pr-readiness-reviewer below. It renders NO merge_ready boolean and NO score; asked for either, it declines and cites the rule.

grain-confidence-reviewer

Surface grain-uniqueness CONFIDENCE as evidence and DIFF two source-map.yaml versions for a Mapping Ready reviewer in the Seshat BI repo. Use when someone asks "is the grain confident?", "what changed in this map?", or is about to review/approve a table's mapping. It READS the already-measured PK-uniqueness signal (PkProof from profile.py / the committed source-profile.md) and reads two map versions from git, then RENDERS a grain-confidence card + a semantic diff and STOPS for the human. "Confidence" here is status + evidence + blockers, NEVER a numeric score. It NEVER approves the gate, NEVER edits source-map.yaml, NEVER picks a new PK/grain, and NEVER auto-resolves a grain/PII/rollup judgment call (Principle V).

kpi-contract-builder

Help a human turn a named-but-planned retail KPI into a governed, owner-ready project metric-contract by driving the shipped kpi_contracts engine (spec 124). On first contact, when no approved decisions exist yet, it assesses source answerability, lists the exact Decision Store decisions to get approved via business-knowledge-interview, and renders a preview with per-field provenance -- writing no YAML. Once the approved kpi_definition / policy_ruling decisions, a named eligible owner, and committed source evidence exist, it drives draft_project_metric_contract (a gold-blocked contract) and, after the gold binding is materialized, finalize_project_metric_contract. Use when someone asks to draft, author, or start a metric contract for a planned KPI in the Seshat BI repo. It never invents business meaning, DAX, SQL, gold columns, approvals, owners, a confidence score, or a registry entry; it never promotes a KPI to Seeded and never self-grants an approval.

pbip-workflow

The git-based workflow for Power BI Project (PBIP) files — saving reports as plain-text TMDL/PBIR, what to commit vs ignore, and the Windows gotchas. Use this whenever someone is putting Power BI under version control: saving a report to git, choosing PBIP vs .pbix, editing TMDL or DAX as text, reviewing a semantic-model change in a PR, or troubleshooting why a .pbi folder, cache, or localSettings is (or isn't) showing up in git. Also use when "Save as PBIP" is missing, when PBIP diffs look noisy (line endings / encoding), or when path-too-long errors appear saving a Power BI project on Windows.

pbip-xray

Audit a committed PBIP semantic model as text (unused fields, relationship risks, measure-graph findings) and explain a TMDL change in business terms for PR review. Read-only: it parses committed TMDL/PBIR and git history, opens no database, never launches Power BI, and edits nothing.

pbir-authoring-adapter

The Power BI report-AUTHORING adapter (F034 completion). Use when someone asks to apply a generated theme to a committed PBIR report, or (later increments) to set visual formatting or a page background by writing the report's PBIR JSON directly -- the settings a human sets by hand in the Power BI UI. This is a COMPANION execution/authoring adapter (the F029-dbt / F030-Dagster pattern), NOT part of the static DEFINE/CHECK core. It writes committed PBIR JSON within a tight allow-list, deterministically and validated; it uses NO pbi-cli, NO live Power BI, NO external dependency, and NO network. It grants no readiness pass and emits no score.

portfolio-watch

Show the recurring, read-only portfolio summary for the Seshat BI repo -- every governed scope's covered-dimension findings (source drift, contract/metric drift, dashboard-intent divergence, readiness, approvals, review), open blockers, human-attention flag, and ONE prioritized next action, plus a baseline diff against the last run (new/resolved/unchanged). Use when someone asks "run portfolio watch", "what changed since last week", "show me the recurring summary", or "which scopes need a human, and what changed". READ-ONLY and invoke-and-present only: it AGGREGATES evidence that shipped surfaces already produce (it runs NO new check, re-derives NOTHING); it persists a LOCAL baseline snapshot under `.seshat/watch/` so the next run can diff; it introduces NO new gate/rule/approval mechanism and emits NO fabricated health/confidence/priority score (every line is a shipped categorical status, a shipped finding enum, or a measured magnitude traced to a committed source).

powerbi-dashboard-design

The Power BI VISUAL/DESIGN layer of the Tower BI Readiness System. Use when someone asks about dashboard design, page layout, a background/canvas asset, a theme/colors choice, a screenshot critique, a mobile layout, or a Power BI handoff in the Seshat BI repo. This skill ROUTES a design request into exactly ONE of four surfaces (report visuals, external background/canvas, theme JSON, implementation handoff) and opens the matching workflow. It is a router and a vocabulary, not a builder: it never edits a PBIP/PBIR file, never generates DAX, and never invents a metric. The gated "design a dashboard from approved contracts" intent it hands off to the F011/012 dashboard-design verb.

pr-readiness-reviewer

Render ONE structured "is this PR safe to merge" verdict for a single pull request in the Seshat_BI repo -- merge_ready (yes/no) + blockers[] + warnings[] + required_human_decisions[] + evidence[] + one next_action. Use when someone asks "is this PR ready to merge?", "what blocks this PR?", or "review this PR for promotion". It OBSERVES the PR's state (open/draft, mergeable, CI/workflow conclusions, open review threads, unresolved Codex/GitHub review comments) and READS the committed readiness evidence the PR touches (readiness-status.yaml, source-map.yaml approval metadata, declared-vs-run tests, no raw data / no secrets / no local paths), then cross-checks the PR body's CLAIMS against that evidence. READ-ONLY: it CANNOT merge a PR, approve a PR, resolve a review thread, push/amend a commit, edit a PR body, or move/upgrade a readiness stage -- it observes and reports only. merge_ready is a DERIVED BOOLEAN, never a numeric merge/confidence/health score (rule #9); a score request is declined.

Data & Documents Listed

readiness-viewer

Show the stage-centric readiness lens across the kit's sources / tables / reports -- one per-stage status MATRIX over the seven readiness stages, each stage's evidence rendered as NAVIGABLE REFERENCES, and an approvals TIMELINE, read from each item's readiness-status.yaml. Use when someone asks "show the readiness viewer", "which stage is each table at", "what evidence backs Gold Ready", or "who approved this gate, and when". READ-ONLY and invoke-and-present only: it RENDERS the recorded state (it never recomputes a status, advances a stage, writes a pass, infers/back-fills an approval, or fabricates evidence), it runs NO validator and opens NO DB connection, and it emits NO numeric health / confidence / percent-ready score (hard rule #9). It is the stage-lens sibling of the F012 control room: same inputs, different view.

release-notes-generator

Draft an evidence-backed per-release note and assess the kit's honest maturity ladder for a shipped roadmap slice in the Seshat BI repo. Use when someone asks to "generate the release notes", "what maturity level is the kit at", "assess the kit's maturity", or "record what became possible in this release". This is Maintenance Automation (F024): it runs as the conductor's post-milestone read, not per a per-run human trigger. It READS the F028 evidence pack + the F032 compatibility matrix + the roadmap ledger, DRAFTS the seven release-note blocks with every capability claim cited, ASSESSES the seven evidence-gated maturity rungs (L0..L6) by a BINARY evidence test each, reports the level as the highest all-evidence-present rung, and then STOPS for the named release owner. It emits NO numeric / maturity / confidence score (the ladder is milestones, not a score), re-measures NOTHING (no seshat check/validate, no DB, no powerbi/ read), self-approves NO release, self-confirms NO level, and publishes nothing.

Data & Documents Listed

report-intent-interview

Given approved metric contracts and a ready semantic model, interview the report requester to capture what a dashboard report is FOR -- audience, purpose, supported decision, review cadence, primary business questions, outcome/driver/guardrail metrics (referenced by name only, never defined), comparisons, dimensions/filters, expected actions/exceptions, pages/drill paths, and mobile/accessibility/language/RTL needs -- and record the result as a committed Report Intent artifact plus a named-human `report_intent_approval` decision. Use when someone describes a report they need in plain language, before any page blueprint, visual spec, or report composition is authored. It asks obvious low-risk items as one batch and critical items individually with explicit named-human approval; it masks PII by default; it records the intent and STOPS. It does NOT define metric meaning, does NOT self-grant any approval, and does NOT emit a confidence score.

retail-build-warehouse

Author the silver + gold migration SQL for a mapped retail table from its APPROVED source-map, then STOP before executing. Use after a table's mapping gate is CLEARED in the Seshat BI repo and someone asks to build the silver/gold SQL, write the migration, or fill the retail-orchestrate build seam. Author-and-stop: it writes warehouse/migrations/*.sql FILES only, never connects to a database, never applies the SQL, and HARD-STOPS at any judgment call the map does not already answer.

retail-control-room

Show the consolidated, cross-table data-quality control room for the Seshat BI repo -- one worst-first roll-up of every table's stage, status, static WARNs, live findings, open blockers, and next action, plus a portfolio-wide open-blockers list. Use when someone asks "show the control room", "what is broken across all tables", or "which table do I fix next". READ-ONLY and invoke-and-present only: it AGGREGATES evidence that already exists (per-table readiness-status.yaml / data-issues.md / blocking-reasons.md + recorded seshat check / retail validate results); it runs NO new validator, edits NO per-table file, clears NO blocker, and emits NO fabricated health/confidence score (every cell is a measured number traceable to a committed source).

retail-discover-portfolio

Discover an unfamiliar retail data portfolio before per-table onboarding. Produces one metadata-only Layer-A survey across every reachable table, proposes non-critical domain and first-delivery scope decisions, delegates selected tables to retail-onboard-table for Layer-B profiling, and hands off to the existing business-knowledge-interview. Never samples source values, self-confirms a decision, or advances beyond the interview handoff.

retail-govern

Run the Seshat BI governance checker and interpret its findings. Use when someone asks to check, validate, or gate Power BI / DAX / TMDL / PBIR / SQL work in the Seshat BI repo, when `seshat check` reports a rule violation, or when you need to know what a rule id (D8, C2, S2, G1, …) means and where to fix it. Invoke-and-interpret only: this skill does NOT build models, run pbi-cli, or auto-fix — it runs the checker and maps ids to fixes.

retail-init

Bootstrap the Compass-Driven kit and lead a NEW user to a first visible result on THEIR own table in the Seshat BI repo. Use when someone asks to "set up the kit", "initialize seshat", "onboard me / get me started", or runs `retail init`. The AGENT performs this: it writes the backstage substrate (compass router + fenced AGENTS.md/CLAUDE.md regions + manifests) via the substrate-only `retail init` CLI, DELEGATES the worked-example offer + human-seam list to `first-hour-compass`, and ROUTES into `retail-onboard-table` for the Stage-1 read-only profile (grain candidates + column types over a live DB, or `[PENDING LIVE PROFILE]` without one). It is NOT a terminal wizard, stores no run-state, fetches no remote, self-grants no approval, and emits no confidence score.

retail-onboard-table

Walk a NEW raw retail table across the FIRST readiness transition -- Source Ready (Stage 1) through Mapping Ready (Stage 2) -- and seed its per-table readiness-status, in the Seshat BI repo. Use when someone says "onboard <schema>.<table>", "take this new table from nothing to a reviewed map", or "start the readiness journey for a table". This is the stage-transition FRONT DOOR: it sequences profile -> map -> gate, writes the readiness-status record, and STOPS at Mapping Ready. It DELEGATES the five mapping artifacts to the `source-mapping` skill (it does not re-implement mapping); it NEVER writes `silver.*` SQL, NEVER self-grants the gate approval, and NEVER answers a Principle-V judgment call (grain, PII, business rollup, product identity).

retail-orchestrate

Drive a retail table through the medallion sequence (profile -> map -> gate -> build -> validate -> Power BI) by composing the existing verb-skills, and self-heal against the governance gate's exit code. Use when someone asks to take a table end-to-end, "orchestrate", "run the whole pipeline", or drive the medallion playbook in the Seshat BI repo. The AGENT is the runtime: this skill sequences and instructs; it never spawns a loop runtime, never self-grants the mapping approval, and HARD-STOPS at human judgment calls.

retail-scaffold

Scaffold the per-table artifact skeletons a retail table needs in the Seshat BI repo. Use when someone asks to create the mapping or readiness files for a new table. It writes empty templates only and answers no judgment call.

retail-semantic-check

Compute the Semantic Model Ready (Stage 5) readiness verdict for a committed Power BI PBIP model in the Seshat BI repo. Use after Gold Ready is `pass`, when someone asks whether the semantic model is ready, to check that every measure binds to an approved metric contract, or before any dashboard/PBIP work. READ-ONLY and invoke-and-interpret only: it runs the existing `seshat check` gate, reads the committed TMDL and the filled metric-contract store (F009), and emits ONE status (not_started | blocked | warning | pass) with evidence + blockers, then STOPS. It NEVER edits TMDL, NEVER defines/approves a metric contract (that is F009), and NEVER calls the Power BI execution adapter (official Power BI MCP / connection; `pbi-cli` no longer preferred) -- that is F016.

retail-validate

Run the LIVE data checks against a materialized retail table and interpret the findings. Use after silver + gold exist for a mapped table in the Seshat BI repo, when someone asks to validate or reconcile a table, or when a V-RC2 / V-RC15 / V-RC16 finding appears. Invoke-and-interpret only: this skill runs `retail validate` against a live Postgres DB and maps each finding id to its fix. It does NOT build models, write SQL, or auto-fix.

run-next-readiness

Advance one table by one readiness stage in the Seshat BI repo: run the next required step, then stop. Use when someone asks what is next for a table and to do it. Never self-grants an approval and never marks a stage pass without its evidence.

showcase-build

Compose a shareable, disclosure-safe, OFFLINE proof bundle of Seshat BI's committed readiness truth -- stages, evidence, blockers, approvals, next actions, metric lineage, a truthful badge/project card, and a four-category disclosure manifest -- for a maintainer or analyst to hand to a reviewer, stakeholder, or prospective contributor. Use when someone asks for a "shareable proof", a "showcase bundle", a "project card", or "something I can send to show what Seshat has governed here". This is a Product Module, artifact-writing: it READS the shipped Explorer projection (`build_explorer_projection`), the Passport (`seshat.passport`), and the disclosure scanner (`seshat.disclosure.scan_disclosure`), composes ONE offline HTML bundle, and STOPS. It recomputes NO readiness, defines NO new evidence schema, invents NO meaning, and grants NO approval. It is fail-closed: a disclosure finding over the FULL composed body blocks generation and writes nothing. Generation is local-only (`.seshat-output/`); publishing the bun

source-mapping

Drive a raw retail source table through the source-mapping gate before any silver SQL exists. Use when someone asks to map, model, profile, or onboard a new bronze table toward Power BI in the Seshat BI repo -- profile the source, decide grain/PK, fill the five mapping artifacts into mappings/<table>/, and stop at the gate. This skill ENFORCES the rule that no silver.* SQL is written until the map is reviewed and approved. It profiles and authors and stops; it does NOT write silver/gold SQL and does NOT build the Power BI model.