cabloy-spec-generation

Generates repository-native PRD, SRS, WBS, test plan, and ADR planning records for Cabloy suites.

971|132|Updated Jan 14, 2018
One-click install
npx skills add https://github.com/cabloy/cabloy --skill cabloy-spec-generation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cabloy-spec-generation
Source: https://github.com/cabloy/cabloy/tree/main/.claude/skills/cabloy-spec-generation
Command: npx skills add https://github.com/cabloy/cabloy --skill cabloy-spec-generation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Long-lived Cabloy business suites need a coherent, traceable planning record set under repo-specs/, but writing PRD, SRS, WBS, test plan, progress, and ADR documents by hand risks inconsistent identifiers, fabricated status, and edition-specific assumptions copied from the wrong repository.

Core Features & Use Cases

  • Edition-aware discovery: Detects Cabloy Basic versus Cabloy Start via repository markers before making site, flavor, UI, or command assumptions.
  • Complete document set generation: Creates README, prd.md, srs.md, pdp-wbs.md, test-plan.md, progress.md, an initial suite-boundary ADR, and derived implementation-gantt.svg and implementation-burndown.svg charts with stable PRD/SRS/WBS/ATP traceability identifiers.
  • Authority and evidence guardrails: Keeps progress as derived status only, refuses verified claims without retained ATP evidence, and routes naming, scaffolding, and contract synchronization to the appropriate sibling skills.
  • Use Case: Ask for a planning baseline for a new training suite with Web and Admin audiences; the skill confirms identity and site strategy, then generates the full repo-specs/a-training/ document set with not-started status and no fabricated evidence.

Quick Start

Ask the assistant to create the complete repo-specs planning set for a new Cabloy suite, providing the providerId, suite name, scope, and audience site strategy.

Frequently Asked Questions about cabloy-spec-generation

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

FAQPage Schema
How do I create a repo-specs planning set for a new Cabloy suite?

Confirm the providerId, suiteName, scope, personas, and Web/Admin site strategy, then the skill generates README, prd.md, srs.md, pdp-wbs.md, test-plan.md, progress.md, an initial ADR, and derived Gantt and burndown SVGs under repo-specs/<suite-short-name>/.

What documents are included in the Cabloy suite spec baseline?

The mandatory baseline is README.md, prd.md, srs.md, pdp-wbs.md, test-plan.md, progress.md, implementation-gantt.svg, implementation-burndown.svg, and decisions/0001-<suite-boundary>.md. Optional records like presentation-contracts.md or provider runbooks are added only when the confirmed scope justifies them.

Does the skill work with both Cabloy Basic and Cabloy Start repositories?

Yes, it detects the edition by checking for __CABLOY_BASIC__ or __CABLOY_START__ markers at the repository root. Edition-specific facts like flavors, sites, and commands are resolved from the active repository and never copied between editions.

Can the progress register mark WBS tasks as verified without test runs?

No. Verified status requires retained ATP evidence with source revision, environment, exact procedure, result, and redacted artifact location. New baselines initialize as not-started, deferred, or blocked, and empty evidence directories are never created.

How are the implementation Gantt and burndown charts kept up to date?

After any change to pdp-wbs.md, test-plan.md, or progress.md, regenerate both SVGs with npm run spec:charts -- <suite> and validate with npm run spec:charts:check -- <suite>. The check fails on stale charts, missing progress rows, or dangling ATP references.

When should planning be routed to other Cabloy skills instead?

Unresolved provider, suite, or module naming goes to cabloy-domain-planning, code generation goes to the backend or frontend scaffold skills, and concrete Vona/Zova contract synchronization goes to cabloy-contract-loop. This skill records handoff points but does not perform implementation.