[PR #5712] docs(skill): unify reverse-engineering and UI/UX workflows #7035

Open
opened 2026-11-01 21:44:53 +00:00 by bsakaguchi · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/5712
Author: @erichroepke
Created: 10/20/2026
Status: 🔄 Open

Base: main ← Head: codex/ui-ux-reverse-engineering


📝 Commits (2)

  • ef384e6 docs(skill): add UI and UX reverse-engineering workflow
  • b2ee1d7 docs(skill): route research reconstruction and verification from one entry point

📊 Changes

6 files changed (+580 additions, -4 deletions)

View changed files

📝 .agents/skills/reverse-engineer-anything/SKILL.md (+60 -3)
📝 .agents/skills/reverse-engineer-anything/references/runtime-observation.md (+5 -0)
➕ .agents/skills/reverse-engineer-anything/references/ui-ux-deconstruction.md (+303 -0)
➕ .agents/skills/reverse-engineer-anything/references/ui-ux-spec-template.md (+100 -0)
➕ .agents/skills/reverse-engineer-anything/references/workflow-routing.md (+111 -0)
📝 src/generatedPackageMetadata.ts (+1 -1)

📄 Description

Summary

Give agents one entry point for product research, software analysis, UI/UX
specification, requested reconstruction and verification. Add outcome routing,
optional specialist guidance, detailed UI/UX inspection and a reusable
specification template, with one evidence index across the selected routes.

Problem and expected behavior

An interface investigation needs more than DOM snapshots: it needs layout and
token extraction, theme/state coverage, measured motion, network/state
correlation, navigation/recovery and accessibility checks. The new workflow
connects those steps to explicit coverage and acceptance criteria, while keeping
observations, declarations, derived measurements, inferences and unknowns distinct.

Change and scope

Add outcome-based routing and progressive UI/UX guidance; cross-link the guides
from the entrypoint and runtime observation. Optional helpers remain optional,
source-only work avoids engine setup, and analysis does not imply a build. Cover
browser, Electron, native and supplied-capture routes without requiring setup for
source-only or screenshot-only tasks. Increment skill version 44 → 46 and
regenerate its package metadata. No issue is linked; this is a focused skill
methodology contribution.

Follow the contribution scope.
Maintainer scope decision: not applicable to this documentation change; it adds
no provider, tool, target execution or injection capability. Optional DevTools/OS
inspection procedures remain subject to available tools and authorized scope.

Contract and boundary impact

  • Semantic owner and earliest changed stage: authored agent skill and references.
  • Contribution scope fit: interface analysis methodology within existing evidence principles.
  • CLI and MCP/tool-catalog contract: unchanged.
  • Provider/platform compatibility: no new capabilities claimed; unsupported routes stay explicit.
  • Evidence/reconstruction contract: source-linked specifications; no promotion of static evidence or visual similarity to runtime equivalence.
  • Target selection/authority/cleanup: existing scope preserved; capture settings and disposable test state restored where used.
  • Generated metadata/package impact: skillVersion changes to 46; ignored bundled skills regenerate through the existing build.

Expected effects:

  • Documentation or usability only; no target interaction.

Evidence and regression coverage

  • Tests added or updated: none; documentation/source/link checks are proportionate to this change.

  • Base evidence: the prior skill had general capture guidance but no UI/UX inspection or blueprint reference.

  • CLI/MCP output: not applicable; no tool behavior changed.

  • Independent review: read-only UI/UX review and six mental routing walkthroughs covered screenshot specifications, source analysis, product/build plans, explicit rebuilds, report-only QA and missing targets. These were not executed runtime trials.

  • Remaining proof gaps: full upstream gates, bundled-skill build and live UI/assistive-technology execution.

  • Observed, derived, and inferred claims remain distinguishable.

  • Artifact identity, source provenance, and failed attempts remain preserved.

  • Unsupported, incomplete, unavailable, or uncertain outcomes remain visible.

Validation performed

  • python3 <skill-creator>/scripts/quick_validate.py .agents/skills/reverse-engineer-anything — passed for source and project copies.
  • Portable-bundle Markdown target/anchor check — 36 checked, zero missing targets. Project aliases and module links were also checked locally and are not part of this upstream diff.
  • node scripts/generate-package-metadata.mjs and node scripts/generate-package-metadata.mjs --check — passed.
  • git diff --cached --check and post-rebase git diff HEAD^ HEAD --check — passed, including new references.
  • Isolated git am of both patch commits — passed against the recorded base; all six resulting files matched the branch byte for byte.
  • npm run check and npm run docs:check — attempted; dependency verification stopped both because the source checkout lacks installed dependencies. No full-gate pass is claimed.

Compatibility, safety, and release

  • Breaking changes/migration: none.
  • Real provider/browser/OS coverage: none added or claimed.
  • Package/release metadata: skill version 46; package version and runtime catalog unchanged.
  • Privacy/containment: private captures and restricted assets stay outside public handoffs; observation does not authorize production mutations or asset-library retrieval.

Author readiness

Human owner: @erichroepke

  • I reviewed the final diff and relevant existing code, understand the change and its edge cases, and take responsibility for the complete PR, including AI-assisted work.
  • I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps.
  • I declare this PR ready for initial maintainer review.

Before merge

  • This PR addresses a concrete problem or an agreed enhancement.
  • Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff.
  • The PR has one focused outcome and the title follows type(scope): outcome.
  • Related issue is linked, or the reason for not linking one is stated above.
  • Tests cover changed observable behavior and meaningful failure paths.
  • Owning docs, contracts, and generated metadata are updated where needed.
  • User-visible CLI/MCP changes include representative output.
  • I checked the final diff for secrets, unrelated cleanup, and unsupported claims.

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/morluto/rea/pull/5712 **Author:** [@erichroepke](https://github.com/erichroepke) **Created:** 10/20/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `codex/ui-ux-reverse-engineering` --- ### 📝 Commits (2) - [`ef384e6`](https://github.com/morluto/rea/commit/ef384e6f313d15948b66b8e674f22e1d6a355b33) docs(skill): add UI and UX reverse-engineering workflow - [`b2ee1d7`](https://github.com/morluto/rea/commit/b2ee1d78905e9ba17149c4d8c59c976c2f85f60a) docs(skill): route research reconstruction and verification from one entry point ### 📊 Changes **6 files changed** (+580 additions, -4 deletions) <details> <summary>View changed files</summary> 📝 `.agents/skills/reverse-engineer-anything/SKILL.md` (+60 -3) 📝 `.agents/skills/reverse-engineer-anything/references/runtime-observation.md` (+5 -0) ➕ `.agents/skills/reverse-engineer-anything/references/ui-ux-deconstruction.md` (+303 -0) ➕ `.agents/skills/reverse-engineer-anything/references/ui-ux-spec-template.md` (+100 -0) ➕ `.agents/skills/reverse-engineer-anything/references/workflow-routing.md` (+111 -0) 📝 `src/generatedPackageMetadata.ts` (+1 -1) </details> ### 📄 Description ## Summary Give agents one entry point for product research, software analysis, UI/UX specification, requested reconstruction and verification. Add outcome routing, optional specialist guidance, detailed UI/UX inspection and a reusable specification template, with one evidence index across the selected routes. ## Problem and expected behavior An interface investigation needs more than DOM snapshots: it needs layout and token extraction, theme/state coverage, measured motion, network/state correlation, navigation/recovery and accessibility checks. The new workflow connects those steps to explicit coverage and acceptance criteria, while keeping observations, declarations, derived measurements, inferences and unknowns distinct. ## Change and scope Add outcome-based routing and progressive UI/UX guidance; cross-link the guides from the entrypoint and runtime observation. Optional helpers remain optional, source-only work avoids engine setup, and analysis does not imply a build. Cover browser, Electron, native and supplied-capture routes without requiring setup for source-only or screenshot-only tasks. Increment skill version 44 → 46 and regenerate its package metadata. No issue is linked; this is a focused skill methodology contribution. Follow the [contribution scope](https://github.com/morluto/rea/blob/main/CONTRIBUTING.md#contribution-scope). Maintainer scope decision: not applicable to this documentation change; it adds no provider, tool, target execution or injection capability. Optional DevTools/OS inspection procedures remain subject to available tools and authorized scope. ## Contract and boundary impact - Semantic owner and earliest changed stage: authored agent skill and references. - Contribution scope fit: interface analysis methodology within existing evidence principles. - CLI and MCP/tool-catalog contract: unchanged. - Provider/platform compatibility: no new capabilities claimed; unsupported routes stay explicit. - Evidence/reconstruction contract: source-linked specifications; no promotion of static evidence or visual similarity to runtime equivalence. - Target selection/authority/cleanup: existing scope preserved; capture settings and disposable test state restored where used. - Generated metadata/package impact: `skillVersion` changes to 46; ignored bundled skills regenerate through the existing build. Expected effects: - [x] Documentation or usability only; no target interaction. ## Evidence and regression coverage - Tests added or updated: none; documentation/source/link checks are proportionate to this change. - Base evidence: the prior skill had general capture guidance but no UI/UX inspection or blueprint reference. - CLI/MCP output: not applicable; no tool behavior changed. - Independent review: read-only UI/UX review and six mental routing walkthroughs covered screenshot specifications, source analysis, product/build plans, explicit rebuilds, report-only QA and missing targets. These were not executed runtime trials. - Remaining proof gaps: full upstream gates, bundled-skill build and live UI/assistive-technology execution. - [x] Observed, derived, and inferred claims remain distinguishable. - [x] Artifact identity, source provenance, and failed attempts remain preserved. - [x] Unsupported, incomplete, unavailable, or uncertain outcomes remain visible. ## Validation performed - `python3 <skill-creator>/scripts/quick_validate.py .agents/skills/reverse-engineer-anything` — passed for source and project copies. - Portable-bundle Markdown target/anchor check — 36 checked, zero missing targets. Project aliases and module links were also checked locally and are not part of this upstream diff. - `node scripts/generate-package-metadata.mjs` and `node scripts/generate-package-metadata.mjs --check` — passed. - `git diff --cached --check` and post-rebase `git diff HEAD^ HEAD --check` — passed, including new references. - Isolated `git am` of both patch commits — passed against the recorded base; all six resulting files matched the branch byte for byte. - `npm run check` and `npm run docs:check` — attempted; dependency verification stopped both because the source checkout lacks installed dependencies. No full-gate pass is claimed. ## Compatibility, safety, and release - Breaking changes/migration: none. - Real provider/browser/OS coverage: none added or claimed. - Package/release metadata: skill version 46; package version and runtime catalog unchanged. - Privacy/containment: private captures and restricted assets stay outside public handoffs; observation does not authorize production mutations or asset-library retrieval. ## Author readiness Human owner: @erichroepke - [ ] I reviewed the final diff and relevant existing code, understand the change and its edge cases, and take responsibility for the complete PR, including AI-assisted work. - [ ] I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps. - [ ] I declare this PR ready for initial maintainer review. ## Before merge - [ ] This PR addresses a concrete problem or an agreed enhancement. - [ ] Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff. - [ ] The PR has one focused outcome and the title follows `type(scope): outcome`. - [ ] Related issue is linked, or the reason for not linking one is stated above. - [ ] Tests cover changed observable behavior and meaningful failure paths. - [ ] Owning docs, contracts, and generated metadata are updated where needed. - [ ] User-visible CLI/MCP changes include representative output. - [ ] I checked the final diff for secrets, unrelated cleanup, and unsupported claims. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
morluto/rea#7035
No description provided.