story

name	story
description	Generate story specifications from MVP stories list. Use when user wants to create detailed story documentation or mentions /story command.

Generate Story Specification

Generate detailed story specifications following the established format, based on MVP stories list and product context.

Usage

/story "Story name"
/story 5                    # By MVP story number
/story                      # Interactive selection

Workflow

Phase 1: Context Gathering

Before generating any specification, read and understand:

1. Product description: ProductSpecification/BriefProductDescription.txt
2. Story mapping: ProductSpecification/stories.md
3. Expected load: ProductSpecification/ExpectedLoad.txt
4. Archived drafts: ProductSpecification/Archived/DraftStories/1st-iteration/
5. Existing specifications: ProductSpecification/stories/*/NN_StoryName.md
6. Story-specific context (optional): ProductSpecification/stories/NN-story-name/interview.md
   • If present, read it for additional context, external documentation, and special instructions

Phase 2: Story Selection

Parse user input to determine target story:

• By name: /story "Login/Logout" — Match story name via ProductSpecification/stories.md
• By number: /story 5 — Story #5 via ProductSpecification/stories.md
• Interactive: /story — List available MVP stories, ask user to choose

Find related archived draft in ProductSpecification/Archived/DraftStories/1st-iteration/ if exists.

Phase 3: Generate Specifications

Load .claude/templates/spec/story-templates.md for document structure.

Generate two files:

• Main spec: NN_StoryName.md (~50 lines max, implementation-focused)
• Notes file: NN_StoryName_Notes.md (warnings, suggestions, technical details)

Phase 4: Hazard Catalogue Scan

Before output, scan the drafted spec against the hazard catalogue — the spec-time, closed-list complement to the open-ended commit-time review passes. Per .claude/guidelines/hazard-catalogue/_index.md (read its "How to apply it"), fan out one hazard-scan-agent per group in the index's Groups list — iterate that list, never a hand-copied set, or a newly-added group goes unchecked — each carrying the drafted spec, _index.md, and its one group file; dispatch them concurrently, collect each pass's GAPs and seam flags, then run one synthesis pass (per _index.md's "Reason across the seams") over the index-named seams and every flagged seam. Fold every GAP back into the spec as an explicit requirement or constraint (and its Notes file) so test-spec and design-preview inherit it. At story altitude the guard is a named requirement, not yet a test — but it must be specific enough that a downstream test could go red on it. An unresolved GAP blocks Phase 5: fold every fired-trigger GAP in, or explicitly dismiss it with a reason, before output.

Phase 5: Output

1. Determine story number from ProductSpecification/stories.md
2. Convert story name to kebab-case for folder name (e.g., "Login/Logout" → "01-login-logout")
3. Create folder if needed: ProductSpecification/stories/NN-story-name/
4. Create both files

Phase 6: Summary

Report: main spec path + line count, notes file path, confirmation, and the hazard-scan result — the group set scanned (the _index.md Groups list at scan time, so a later group addition can re-trigger per _index.md's "A new group obligates a re-scan"), each group's verdict, and every GAP's disposition (folded → named requirement, or dismissed with reason).

Design Constraints

• Language: English
• Main file brevity: Target ~50 lines max — ruthlessly cut fluff
• Notes completeness: All warnings, suggestions, technical details go to Notes file
• Archived drafts: Use as reference but apply new compact format
• No redundancy: If it's in main file, don't repeat in notes
• Check if spec already exists before creating (avoid duplicates)