stand-scenarios

Converts product scenario documents into testable claims with oracles, layers, and priorities.

10|9|Updated May 22, 2026
One-click install
npx skills add https://github.com/constructorfabric/insight --skill stand-scenarios-constructorfabric
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stand-scenarios
Source: https://github.com/constructorfabric/insight/tree/main/.claude/skills/stand-scenarios
Command: npx skills add https://github.com/constructorfabric/insight --skill stand-scenarios-constructorfabric

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Product requirement documents describe boundaries and invariants in prose, leaving QA engineers to guess which guarantees are testable, which surfaces exist, and what oracle to assert against. This Skill turns a scenario document into concrete, prioritized test claims grounded in seeded fixtures and actual API routes. ## Core Features & Use Cases - Claim design from product docs: Maps personas (EXEC/LEAD/IC/ADMIN) to seeded fixtures, checks whether each surface exists, and emits claims with an oracle, layer (API vs UI), and blast-radius priority (P0/P1/P2). - Coverage triage: Records surface verdicts (Built/Partial/No surface/Not enforceable here), attributes existing tests to product clauses, and tracks issued claim IDs in SCENARIO-COVERAGE.md. - Findings capture: Surfaces divergences between the product document and the implementation, such as two group-size thresholds disagreeing on the number. - Use Case: Ask which product rules for scenario S-9 are unverified, and receive designed claims like "a lead asking for their own manager's subchart is refused, 404 not 403" with the oracle, route, priority, and confidence tag ready for test implementation. ## Quick Start Ask the assistant to design testable claims for scenario S-9 from the product scenarios document, grounded in the seeded roster.

Frequently Asked Questions about stand-scenarios

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

FAQPage Schema
How do I turn a product requirements document into test cases?

Pick one scenario, persona, or invariant, then answer five questions in order: which seeded fixture maps to the persona, whether the surface exists, what oracle to assert against, which layer (API preferred over UI), and what already tests it. Emit each claim with an ID, priority, and confidence tag.

What makes a good negative test for access control boundaries?

Target the 'never' rules in the requirements, since reach bugs are silent and no user reports seeing too much. Assert the whole partition in one request (visible, out-of-scope, out-of-tenant, unresolvable ids) and verify the correct refusal code, since 404 and 403 leak different information.

Should boundary tests run at the API or UI layer?

Prefer the API unless the claim can only be true in a browser. A boundary enforced only by the frontend is the defect the rule exists to prevent, and browser tests are slower and flakier than HTTP calls proving the same thing.

Can I assert exact metric values in scenario tests?

No. Reading a number off a running stand and asserting it back only proves the code produced itself. Assert shape, reach, presence, honesty, and refusal instead, deriving expectations from the manifest or the response at runtime.

What should I do when a scenario has no built surface to test?

Record a 'No surface' verdict in one line and stop rather than designing a test that would pass for the wrong reason. If the rule is real but belongs to another suite, name the owning suite instead of filing it as absent.