story-done

Verify design story completion against acceptance criteria, evidence, and registries.

Updated Apr 25, 2026
One-click install
npx skills add https://github.com/easonlao/jiazi-game --skill story-done-easonlao
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: story-done
Source: https://github.com/easonlao/jiazi-game/tree/main/.claude/skills/story-done
Command: npx skills add https://github.com/easonlao/jiazi-game --skill story-done-easonlao

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents stories from being marked complete when acceptance criteria, evidence, and design/architecture alignment have not actually been verified, reducing drift between GDD/ADR intent and implementation reality.

Core Features & Use Cases

  • Acceptance-criteria verification: Reads a story file and checks each acceptance criterion via file/test/string and dependency checks, with optional manual confirmation for subjective or hard-to-automate items.
  • Traceability and evidence gating: Builds a criterion-to-test traceability table and enforces evidence expectations based on story type (e.g., BLOCKING for missing Logic unit tests, advisory for missing UI/Visual evidence).
  • Deviation detection with explicit reporting: Checks for GDD/registry mismatches, manifest staleness, ADR constraint issues, numeric hardcoding, and out-of-scope file touches, then prompts for code review and surfaces the next ready story.

Quick Start

Run the story-done skill with the target story file path to validate acceptance criteria, document deviations, and mark the story as Complete.

Frequently Asked Questions about story-done

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

FAQPage Schema
How do I verify acceptance criteria before marking a story as complete?

To verify acceptance criteria for story completion, cross-check each criterion against required test evidence, registries, and manifests. The process enforces traceability and runs full QA and code-review gates before updating the story status to Complete.

What is ADR deviation checking in a sprint workflow?

ADR deviation checking detects mismatches between implementation reality and architectural design records. It identifies GDD/registry mismatches, manifest staleness, numeric hardcoding, and out-of-scope file touches to prevent drift from original architecture intent during story closure.

How do I enforce test evidence requirements for logic and UI stories?

Enforce test evidence requirements by building a criterion-to-test traceability table that applies blocking gates for missing Logic unit tests and advisory checks for missing UI/Visual evidence. This ensures deterministic verification before a story is marked complete.

Does story closure verification work without manual confirmation?

Story closure verification performs automated file, test, string, and dependency checks automatically, but requires optional manual confirmation for subjective or hard-to-automate acceptance criteria. Full mode also requires passing QA and code-review gates.

What happens when GDD and registry mismatches are detected during story completion?

When GDD and registry mismatches are detected during story completion, the process reports deviations explicitly, prompts for code review, and surfaces the next ready story. The story status is not updated to Complete until all blocking issues are resolved.

Best way to close the loop on an implementation story and reveal next sprint work?

The best way to close the loop on an implementation story is to run an end-of-implementation closure workflow that cross-checks acceptance criteria, validates test evidence, and detects ADR deviations. This deterministic verification updates story status and reveals the next sprint work.