Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .claude-plugin/marketplace.json
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@
"name": "superloopy",
"source": "./",
"description": "Strict evidence loop harness: repo-local loop state, success criteria, append-only ledgers, evidence-gated completion, plus skills (loop, research, frontend, clone) and a read-only/worker crew. Works on Claude Code and Codex from one repo.",
"version": "0.11.1",
"version": "0.12.0",
"license": "MIT",
"keywords": [
"claude-code",
Expand Down
2 changes: 1 addition & 1 deletion .claude-plugin/plugin.json
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "superloopy",
"version": "0.11.1",
"version": "0.12.0",
"description": "Lightweight loop harness with strict evidence gates — Claude Code edition.",
"homepage": "https://github.com/beefiker/superloopy#readme",
"repository": "https://github.com/beefiker/superloopy",
Expand Down
2 changes: 1 addition & 1 deletion .codex-plugin/plugin.json
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "superloopy",
"version": "0.11.1",
"version": "0.12.0",
"description": "Lightweight loop harness with strict evidence gates.",
"homepage": "https://github.com/beefiker/superloopy#readme",
"repository": "https://github.com/beefiker/superloopy",
Expand Down
2 changes: 1 addition & 1 deletion docs/superloopy-design-audit.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@ This doctor-verified audit records Superloopy's own design decisions. It exists
| `frontend-quality-overlays` | StyleGallery is unrelated to the user-reported UX fixes already owned by the shared frontend base, but its architecture research exposed independent quality gaps in spatial ownership, temporal review, visual-reference authority, high-impact rationale, and bounded debt. | Superloopy adds a StyleGallery-informed but independently authored quality overlay: target routing remains unchanged; `layout.md`, `motion-core.md`, and visual-reference discipline activate only from changed claims; rationale, task-based usability evidence, and accepted debt scale by consequence; Web implementation guidance remains a specialization; and campaign heuristics cannot become unsupported universal blockers. Spatial disclosure uses target-derived exposure modes with generic source/summary and disclosed/destination state proof; collapsed/expanded proof applies only when that mechanism is selected. The overlay rejects fixed Web breakpoint matrices, CSS pattern catalogs, mandatory image generation, and universal motion values as cross-platform defaults. It does not add motion or haptic artifact kinds because file existence is not proof of temporal behavior or physical output; a future performance claim should wait for a machine-validated structured report tied to raw-trace hash, target, build, metric, and threshold. The reviewed StyleGallery revision is pinned only as research provenance because no root license was identified, so no code, prose, schema, pattern, generated file, or dependency is copied. | `skills/superloopy-frontend/SKILL.md`, `skills/superloopy-frontend/references/layout.md`, `skills/superloopy-frontend/references/motion-core.md`, `skills/superloopy-frontend/references/motion.md`, `skills/superloopy-frontend/references/image-first.md`, `skills/superloopy-frontend/references/anti-slop.md`, `skills/superloopy-frontend/references/ux.md`, `skills/superloopy-frontend/references/upstream-notice.md`, `test/frontend-quality-overlays.test.js`, and the file-audit/golden-set inventories. |
| `frontend-context-routing` | Existing Web guidance had no bounded lane for verified platform design-system contracts, implementation-level scroll motion, or redesigns that must preserve a living product. | The Web route classifies product/application DOM, marketing/editorial, PWA/extension, custom-rendered, and embedded-client delivery before loading optional references or selecting evidence. SEO is conditional on the current Web target being public and crawlable; native or embedded work scores only a distinct public Web deployment. Design-system and brand references remain subordinate to the repository's authoritative source and express color, type, spacing, component states, and motion in project-native tokens, units, and APIs rather than imposing CSS/hex/4px conventions on every target; image-first and motion routes activate only for affected visual/motion claims and derive proof from the target. System adoption requires an actual contract, current official-doc/package verification, and dependency approval; approved implementation executes one verified package-manager command and validates integration, while advice only presents it. Redesigns mark unavailable facts unverified, preserve IA by default, and change broken IA only when separately in scope and explicitly approved. Selected mechanisms are adapted from Taste Skill under MIT with the full notice packaged locally. | `skills/superloopy-frontend/references/design-system.md`, `skills/superloopy-frontend/references/design/_INDEX.md`, `skills/superloopy-frontend/references/image-first.md`, `skills/superloopy-frontend/references/web.md`, `skills/superloopy-frontend/references/system-map.md`, `skills/superloopy-frontend/references/motion.md`, `skills/superloopy-frontend/references/redesign.md`, `skills/superloopy-frontend/references/upstream-notice.md`, and `test/plugin.test.js`. |
| `korean-humanizer-skill` | Korean users need a Codex-native AI-tone-removal workflow with stronger proof than a prompt-only port. | Superloopy ships a local `humanize-korean` skill with packaged references, attribution, deterministic audit metrics, and evidence artifacts. | `skills/humanize-korean/SKILL.md`, `skills/humanize-korean/scripts/audit-humanize-output.mjs`, `test/humanize-korean.test.js`, `test/plugin.test.js`, and the file-audit/golden-set inventories. |
| `slides-skill` | Presentation work needs the same show-don't-tell style discovery and rendered-screenshot proof as other visual lanes, and rebuilding a mature template library from scratch would be weaker than adapting one. | A bundled `superloopy-slides` skill generates zero-dependency single-file HTML decks on a fixed 16:9 stage: three rendered style previews the user picks from, a vendored MIT `frontend-slides` template pack (audited by provenance and exempt from the line cap, like the landing-page orbit bundles), and completion gated on rendered-screenshot visual QA under `.superloopy/evidence/slides/`. | `skills/superloopy-slides/SKILL.md`, `skills/superloopy-slides/LICENSE`, `test/plugin.test.js`, `test/doctor.test.js`, and the file-audit/golden-set inventories. |
| `slides-skill` | Presentation work needs the same show-don't-tell style discovery and rendered-screenshot proof as other visual lanes, and rebuilding a mature template library from scratch would be weaker than adapting one. | A bundled `superloopy-slides` skill generates zero-dependency single-file HTML decks on a fixed 16:9 stage: three rendered style previews the user picks from (a user-named brand may inform one slot via a single frontend StyleGallery teardown as scoped reference input), a vendored MIT `frontend-slides` template pack (audited by provenance and exempt from the line cap, like the landing-page orbit bundles), and completion gated on rendered-screenshot visual QA under `.superloopy/evidence/slides/` — the run root created via the shared portable evidence helper's slides lane, with the frontend anti-slop non-waivable core and motion floor binding every deck. | `skills/superloopy-slides/SKILL.md`, `skills/superloopy-slides/LICENSE`, `test/plugin.test.js`, `test/doctor.test.js`, and the file-audit/golden-set inventories. |
| `dual-host-support` | Superloopy's core is host-agnostic and users want it on Claude Code as well as Codex. | A parallel `.claude-plugin/` manifest + marketplace, `hooks/hooks.json`, `agents/*.md`, and a Claude model-policy doc let one repo install on both hosts; a shared `src/receipt.js` recovers the SubagentStop receipt from last_assistant_message or the subagent transcript tail, and `isClaudeHost` makes bootstrap a no-op on Claude — the deterministic floor stays host-agnostic. | `.claude-plugin/plugin.json`, `src/receipt.js`, `src/agents.js`, `test/host-adapter.test.js`, and the file-audit/golden-set inventories. |

## Compatibility Boundary
Expand Down
4 changes: 2 additions & 2 deletions docs/superloopy-file-audit.md
Original file line number Diff line number Diff line change
Expand Up @@ -329,14 +329,14 @@ Superloopy is its own lightweight loop harness: one small CLI, repo-local `.supe
| `skills/superloopy-frontend/references/qt-quick.md` | Qt Quick/QML contract for one Controls style route, module and geometry/state ownership, accessibility and tooling, target-specific desktop versus mobile transitions, and mixed QQuickWidget boundaries. | Superloopy-native; original prose grounded in linked official [Qt Quick Controls style documentation](https://doc.qt.io/qt-6/qtquickcontrols-styles.html) and related Qt Quick references; no external source files vendored. |
| `skills/superloopy-frontend/references/qt-qa.md` | Claim- and risk-scoped Qt command, state, behavioral, optional visual-capture, screenshot-interpretation, and unverified-surface disclosure gate. | Superloopy-native; original prose grounded in linked official [Qt Test documentation](https://doc.qt.io/qt-6/qttest-best-practices.html) and related Qt tooling references; a visual artifact is required only for changed visual or visibly consequential interaction claims, and browser/offscreen proof cannot substitute for native evidence. |
| `skills/superloopy-frontend/scripts/ds-compliance.mjs` | Dependency-free partial token lint for undeclared 3/4/6/8-digit hex and off-scale integer, fractional, or negative px values on exact CSS spacing properties, including multiline declarations, while allowing absolute 0/1px and ignoring border-side lookalikes. | Superloopy-native Node-built-in helper with value-line reporting and a Windows-safe CLI entry guard; non-zero violations remain bounded color/spacing proof, not typography, semantics, accessibility, or design-system adoption. |
| `skills/superloopy-frontend/scripts/evidence-root.mjs` | Dependency-free cross-shell helper that creates one portable timestamped frontend evidence root and verifies non-empty regular artifacts confined beneath it. | Superloopy-native Node-built-in helper; rejects invalid or Windows-reserved slugs, duplicate run roots, traversal, symlink ancestors/leaves, empty files, and out-of-root evidence without adding a project dependency. |
| `skills/superloopy-frontend/scripts/evidence-root.mjs` | Dependency-free cross-shell helper that creates one portable timestamped frontend or slides evidence root and verifies non-empty regular artifacts confined beneath it. | Superloopy-native Node-built-in helper; rejects invalid or Windows-reserved slugs, duplicate run roots, traversal, symlink ancestors/leaves, empty files, and out-of-root evidence without adding a project dependency. |
| `skills/superloopy-frontend/scripts/visual-diff.mjs` | Dependency-free PNG visual-diff for the frontend visual-QA gate: similarityScore, alpha intactness, and ranked grid hotspots as review evidence. | Superloopy-native; uses only node:zlib/node:fs, no external image library or vendored code. |
| `skills/superloopy-loop/SKILL.md` | Skill instructions for Superloopy guide, proof, capture, evidence, check, finish, review, checkpoint, and doctor flow. | Uses Superloopy-native workflow terms. |
| `skills/superloopy-loop/agents/openai.yaml` | Minimal agent metadata for Superloopy discovery. | Search terms are Superloopy-native. |
| `skills/superloopy-research/SKILL.md` | Explicit-invocation-gated skill instructions for deep research with EXPAND waves, claim verification, cited synthesis, and Superloopy evidence artifacts. | Superloopy-native research workflow; workers return text and parent-owned artifacts remain authoritative. |
| `skills/superloopy-research/agents/openai.yaml` | Minimal explicit-only agent metadata for discovering the Superloopy research skill without implicit activation. | Superloopy-native skill metadata only. |
| `skills/superloopy-slides/LICENSE` | Upstream MIT license for the adapted frontend-slides skill content. | License text only; records third-party provenance and no executable plugin logic. |
| `skills/superloopy-slides/SKILL.md` | Skill instructions for zero-dependency HTML slide decks: style discovery, fixed 16:9 stage, PPT conversion, PDF export, and a visual-QA evidence gate. | Superloopy-native evidence gate on `.superloopy/evidence/slides/`; adapted from the MIT frontend-slides skill with attribution. |
| `skills/superloopy-slides/SKILL.md` | Skill instructions for zero-dependency HTML slide decks: style discovery with an optional brand-gallery route, fixed 16:9 stage, PPT conversion, PDF export, the frontend anti-slop core, and a visual-QA evidence gate on the shared portable evidence helper. | Superloopy-native evidence gate on `.superloopy/evidence/slides/`; adapted from the MIT frontend-slides skill with attribution. |
| `skills/superloopy-slides/STYLE_PRESETS.md` | Twelve curated slide style presets with palettes, fonts, and signature elements. | Design reference content for deck generation only; no executable plugin logic. |
| `skills/superloopy-slides/agents/openai.yaml` | Minimal agent metadata for discovering the Superloopy slides skill. | Superloopy-native skill metadata only. |
| `skills/superloopy-slides/animation-patterns.md` | CSS/JS animation snippets and effect-to-feeling guide for slide generation. | Design reference content for deck generation only; no executable plugin logic. |
Expand Down
4 changes: 2 additions & 2 deletions docs/superloopy-loop-golden-set.md
Original file line number Diff line number Diff line change
Expand Up @@ -349,14 +349,14 @@ Total: 100 points.
| `skills/superloopy-frontend/references/qt-quick.md` | Original-prose contract; `test/frontend-qt-contract.test.js` and official [Qt Quick Controls style documentation](https://doc.qt.io/qt-6/qtquickcontrols-styles.html). | Must select one Controls style route before loading QML, keep module/layout/state ownership singular, preserve accessibility, use desktop resize/mixed-DPR checks only where exposed, use mobile window/orientation/posture/inset transitions on mobile/tablet, and disclose QQuickWidget limits. |
| `skills/superloopy-frontend/references/qt-qa.md` | Original-prose contract; `test/frontend-qt-contract.test.js` and official [Qt Test documentation](https://doc.qt.io/qt-6/qttest-best-practices.html). | Must classify repository gates, require executable builds and real-target behavioral proof, select states by changed journey/risk, require native capture only for changed visual or visibly consequential interaction claims, retain human screenshot judgment, disclose unverified surfaces, and never accept browser/offscreen proof as native evidence. |
| `skills/superloopy-frontend/scripts/ds-compliance.mjs` | `test/ds-compliance.test.js` and `test/plugin.test.js`. | Must remain dependency-free, normalize 3/4/6/8-digit hex, flag undeclared colors and off-scale integer/fractional/negative values only on exact CSS spacing properties including multiline declarations, report the value line, ignore border-side lookalikes, avoid partial px matches, allow absolute 0/1px, invoke correctly on Windows, and exit non-zero on violations without claiming broader design compliance. |
| `skills/superloopy-frontend/scripts/evidence-root.mjs` | `test/frontend-routing-scenarios.test.js`, `test/plugin.test.js`, and package inventory. | Must create one portable `.superloopy/evidence/frontend/YYYYMMDDTHHMMSSZ-<slug>` root from an external target project and verify non-empty regular artifacts while rejecting reserved/unsafe slugs, duplicate roots, traversal, symlink paths, empty files, and out-of-root proof without dependencies. |
| `skills/superloopy-frontend/scripts/evidence-root.mjs` | `test/frontend-routing-scenarios.test.js`, `test/plugin.test.js`, and package inventory. | Must create one portable `.superloopy/evidence/<frontend|slides>/YYYYMMDDTHHMMSSZ-<slug>` root from an external target project and verify non-empty regular artifacts while rejecting unknown lanes, reserved/unsafe slugs, duplicate roots, traversal, symlink paths, empty files, and out-of-root proof without dependencies. |
| `skills/superloopy-frontend/scripts/visual-diff.mjs` | `test/visual-diff.test.js`. | Must decode 8-bit PNGs and emit dimensionsMatch, diffRatio, similarityScore, alphaChannelIntact, and ranked grid hotspots with no dependencies. |
| `skills/superloopy-loop/SKILL.md` | `test/docs.test.js`, doctor skill check. | Must describe guide, proof, capture, evidence, check, finish, gates, and receipt rules accurately. |
| `skills/superloopy-loop/agents/openai.yaml` | Audit coverage and reviewability check. | Must remain minimal Superloopy discovery metadata. |
| `skills/superloopy-research/SKILL.md` | `test/plugin.test.js`, audit coverage. | Must require explicit invocation (`$superloopy:superloopy-research`, `/superloopy:superloopy-research`, or a leading `loopy`/`루피`), reject plain research/investigate/summarize vocabulary as activation, and describe exhaustive deep research with EXPAND waves, claim verification, cited synthesis, and Superloopy evidence receipts. |
| `skills/superloopy-research/agents/openai.yaml` | Audit coverage and reviewability check. | Must remain minimal explicit-only Superloopy discovery metadata for deep research. |
| `skills/superloopy-slides/LICENSE` | Audit coverage and reviewability check. | Must remain Superloopy-native slides skill support content with no plugin runtime coupling. |
| `skills/superloopy-slides/SKILL.md` | Audit coverage and reviewability check. | Must gate deck completion on a real-browser visual-QA artifact under `.superloopy/evidence/slides/` with a `SUPERLOOPY_EVIDENCE` receipt. |
| `skills/superloopy-slides/SKILL.md` | `test/plugin.test.js`, audit coverage, and reviewability check. | Must gate deck completion on a real-browser visual-QA artifact under `.superloopy/evidence/slides/` (created via the shared portable evidence helper) with a `SUPERLOOPY_EVIDENCE` receipt, bind the frontend anti-slop non-waivable core, and open the brand gallery only for a user-named brand or mood as scoped reference input. |
| `skills/superloopy-slides/STYLE_PRESETS.md` | Audit coverage and reviewability check. | Must remain Superloopy-native slides skill support content with no plugin runtime coupling. |
| `skills/superloopy-slides/agents/openai.yaml` | Audit coverage and reviewability check. | Must remain Superloopy-native slides skill support content with no plugin runtime coupling. |
| `skills/superloopy-slides/animation-patterns.md` | Audit coverage and reviewability check. | Must remain Superloopy-native slides skill support content with no plugin runtime coupling. |
Expand Down
4 changes: 2 additions & 2 deletions package-lock.json

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion package.json
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "superloopy",
"version": "0.11.1",
"version": "0.12.0",
"description": "A lightweight loop harness for Codex and Claude Code with strict evidence gates.",
"type": "module",
"bin": {
Expand Down
Loading
Loading