retail-scaffold

Scaffolds new seshat check rule stubs and verifies five-place wiring for governance rules.

2|Updated Jun 22, 2026
One-click install
npx skills add https://github.com/Kemetra/Seshat-BI --skill retail-scaffold-kemetra
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: retail-scaffold
Source: https://github.com/Kemetra/Seshat-BI/tree/main/.claude/skills/retail-scaffold
Command: npx skills add https://github.com/Kemetra/Seshat-BI --skill retail-scaffold-kemetra

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Adding a new seshat check governance rule requires wiring the same five places every time, and doing it by hand silently under-governs a rule when a place is missed. This Skill runs the retail scaffold CLI verb to automate the mechanical authoring work and to detect wiring drift in existing rules. ## Core Features & Use Cases - Author mode: Runs retail scaffold --id <RULE_ID> --title "<title>" to write a stub rule module, a failing test stub, and the EXPECTED_RULE_IDS entry, while printing golden-record regen commands and glossary edits for a human to apply. - Doctor mode: Runs retail scaffold --doctor to sweep every registered rule id and report which of the five wiring places each is present in or missing from, exiting non-zero on drift. - Use Case: When adding a new data-quality rule to the Seshat BI retail pipeline, invoke this Skill to generate the honest-red stub and wiring, then fill in the real check logic and let the test suite confirm the rule is fully wired. ## Quick Start Ask the assistant to scaffold a new seshat check rule with a given rule id and title, or to run a doctor sweep verifying that all registered rules are wired into every required place.

Frequently Asked Questions about retail-scaffold

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add a new seshat check rule?▼

Run retail scaffold with the rule id and a one-line title in author mode. It writes a stub rule module, a failing test stub, and the EXPECTED_RULE_IDS entry, then prints the remaining manual steps such as golden-record regeneration and the glossary row.

How do I check whether a governance rule is fully wired?▼

Use doctor mode with retail scaffold --doctor to sweep every registered rule id, or pass --id to verify one rule. It reports which of the five wiring places each rule is present in or missing from and exits non-zero on any drift.

What is the difference between retail-scaffold and retail-govern?▼

retail-scaffold authors new rules and verifies their wiring, while retail-govern interprets existing seshat check findings by mapping a reported rule id to its meaning and fix. Use scaffold when adding a rule and govern when a check has failed.

Does the scaffold write the glossary entry or golden records automatically?▼

No. The helper writes exactly three targets and prints the glossary row suggestion, golden-record regen commands, and the import edit for a human to apply by hand. This print-only split keeps prose and golden records under human control.

Why does the generated test stub fail on purpose?▼

The stub test fails deliberately as an honest red signal that the rule contains no real logic yet. The author must replace it with a genuine test after filling in the check implementation, ensuring no rule ships as an empty no-op.