Skip to content

feat: improve wardley command with math-validated stage boundaries#191

Closed
tractorjuice wants to merge 1 commit intomainfrom
feat/improve-wardley-prompt
Closed

feat: improve wardley command with math-validated stage boundaries#191
tractorjuice wants to merge 1 commit intomainfrom
feat/improve-wardley-prompt

Conversation

@tractorjuice
Copy link
Copy Markdown
Owner

Summary

  • Fixed stage boundary ranges to match validate-wardley-math.mjs — Genesis 0.00-0.24, Custom 0.25-0.49, Product 0.50-0.74, Commodity 0.75-1.00
  • Added avoid-boundary-values rule — use 0.72 not 0.75 to prevent stage ambiguity
  • Added OWM-to-table consistency requirement — map code coordinates must match Component Inventory
  • Added positioning rationale instruction — explain why each component sits at its evolution stage
  • Added GOV.UK Design System example — common edge case at the Product/Commodity boundary

How this was found

Used the autoresearch optimiser with wardley math validation added to the structural gate:

Iteration Score Status Issue
0 (baseline) FAIL discard GOV.UK Design System evo 0.78 labelled Product (Commodity)
1 FAIL discard GOV.UK Design System evo 0.75 labelled Product (boundary)
2 9.0 keep Avoid boundary values + positioning rationale

The math validation caught bugs that LLM-as-judge scoring missed.

Test plan

  • Run /arckit:wardley and verify no stage-evolution mismatches
  • Verify no exact boundary values (0.25, 0.50, 0.75) in output
  • Verify OWM coordinates match Component Inventory table
  • Run python scripts/converter.py to regenerate extension formats

🤖 Generated with Claude Code

Fix stage boundary ranges to match validate-wardley-math.mjs
(Genesis 0.00-0.24, Custom 0.25-0.49, Product 0.50-0.74,
Commodity 0.75-1.00). Add avoid-boundary-values rule, OWM-to-table
consistency requirement, and positioning rationale instruction.

Found via autoresearch — baseline and first iteration both failed
structural validation due to stage-evolution mismatches that
LLM-as-judge scoring missed entirely.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
tractorjuice added a commit that referenced this pull request Mar 27, 2026
Merged 6 pre-autoresearch prompt quality PRs into the combined branch:

- #187 requirements: user personas, UC structure, Given/When/Then AC
- #188 adr: measurable consequences with project-specific metrics
- #189 backlog: strict acceptance criteria rules (testable, thresholds)
- #191 wardley: math-validated stage boundaries table
- #192 wardley.value-chain: visibility rationale and validation checklist
- #194 wardley.climate: recommended action field with specificity rules

All autoresearch additions (scores, traceability) preserved alongside
the quality improvements. Lint passes. Converter regenerated.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@tractorjuice
Copy link
Copy Markdown
Owner Author

Merged into #265 (combined autoresearch PR)

@tractorjuice tractorjuice deleted the feat/improve-wardley-prompt branch March 27, 2026 23:52
tractorjuice added a commit that referenced this pull request Apr 6, 2026
Merged 6 pre-autoresearch prompt quality PRs into the combined branch:

- #187 requirements: user personas, UC structure, Given/When/Then AC
- #188 adr: measurable consequences with project-specific metrics
- #189 backlog: strict acceptance criteria rules (testable, thresholds)
- #191 wardley: math-validated stage boundaries table
- #192 wardley.value-chain: visibility rationale and validation checklist
- #194 wardley.climate: recommended action field with specificity rules

All autoresearch additions (scores, traceability) preserved alongside
the quality improvements. Lint passes. Converter regenerated.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant