markilo 1.0.0 → 1.1.2

package/README.md

@@ -1,349 +1,133 @@
 <p align="center">
   <picture>
-    <source media="(prefers-color-scheme: dark)" srcset="assets/logo-dark.svg">
-    <img src="assets/logo-light.svg" alt="Markilo" width="360">
+    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/logo-dark.svg">
+    <img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/logo-light.svg" alt="Markilo" width="340">
   </picture>
 </p>
 
+<h3 align="center">A brand identity studio inside Claude Code.</h3>
+
 <p align="center">
-  <b>The open-source brand identity studio for Claude Code.</b><br>
-  Research a brand, design original SVG identities, check them against the marks and names you collected, and export every asset you need.
+  Three original identities for your product, drawn as real SVG and checked for originality.<br>
+  You compare them, choose or ask for changes — and get every file the product needs.
 </p>
 
 <p align="center">
   <a href="https://github.com/cihantopac/markilo/actions/workflows/ci.yml"><img alt="CI" src="https://github.com/cihantopac/markilo/actions/workflows/ci.yml/badge.svg"></a>
-  <a href="LICENSE"><img alt="License: Apache-2.0" src="https://img.shields.io/badge/license-Apache--2.0-d12c12"></a>
-  <img alt="Version" src="https://img.shields.io/badge/version-1.0.0-101218">
-  <img alt="Platforms" src="https://img.shields.io/badge/platforms-macOS%20%7C%20Linux%20%7C%20Windows-101218">
+  <a href="https://www.npmjs.com/package/markilo"><img alt="npm" src="https://img.shields.io/npm/v/markilo?color=7e1d2a&label=npm"></a>
+  <a href="https://github.com/cihantopac/markilo/blob/main/LICENSE"><img alt="License: Apache-2.0" src="https://img.shields.io/badge/license-Apache--2.0-16120f"></a>
 </p>
 
----
-
-Markilo is not a "generate a logo" prompt. It is a Claude Code plugin plus a deterministic toolchain that
-**researches** a product, **designs** original vector marks, **measures** them (small-size legibility,
-balance, contrast, one-color reproduction), **compares** them against competitors and each other, and
-**exports** a complete, platform-correct asset kit — all locally, with no servers, paid APIs or accounts.
-
-![Markilo's own identity: every logo file, exported by Markilo](assets/kit-overview.png)
-
-<sub>Markilo's identity was made by running Markilo on itself, end to end: repository and market research (64 sources,
-37 competitor and category marks), five new directions plus the previous identity, an early sketch review, an
-independent critic gate that only one concept passed (`overlap`: two arches sharing a vermilion stem), then finalize
-and export. Every image in this README and the plugin icon is an unedited Markilo export (`branding/`). No trademark
-registry search has been done yet; the guidelines say so.</sub>
-
-## Highlights
-
-- **Brand intelligence, with receipts.** Repository and website scans, competitor and market research, and a persistent
-  Brand Intelligence Profile where every finding is `observed`, `inferred`, `verified` or `user-provided` and carries a
-  source and a date. Markilo reads what the product actually does — value proposition, features, audience, surfaces,
-  platforms, languages, pricing — and keeps what the product states apart from what it inferred. Unknowns are recorded
-  as gaps instead of guesses.
-- **Originality engine.** Silhouette overlap under rotation and mirroring, perceptual hashes (aHash/dHash/DCT pHash),
-  structural descriptors, palette distance (CIEDE2000), stock-primitive detection, concept-to-concept diversity, name
-  confusion (spelling, sound, stems) and trademark pre-screening plans for WIPO, TMview, USPTO, TÜRKPATENT and UKIPO —
-  written up as an Originality & Similarity Report with side-by-side overlays. Verdicts say how much evidence they rest
-  on — counted in distinct brands, with official, store and catalog sources rated apart: with too few references or no
-  logged registry searches the answer is `insufficient-evidence`, never "clear". Calibrated on 24 real brands' logos
-  (AUC 0.99; mirrored, rotated and recolored copies always caught; see [docs/ORIGINALITY.md](docs/ORIGINALITY.md)). It
-  finds conflicts among what was collected; it cannot prove a mark is unique or registrable.
-- **Ideas before drawings.** Every identity starts from a longlist of at least 12 visual hooks from the product, the
-  audience, the name, type and abstract character; a critic strikes the ones a stranger would read as a UI glyph or a
-  cliché from the words alone, before anything is drawn, and the gallery shows what was kept and why.
-- **Real design output.** Symbols built from construction specs — primitives, booleans, gaps, offsets, fillets,
-  repeats and real letterforms, compiled with Skia's path engine into clean SVG — and tuned by measured variants.
-  Wordmarks outlined from 24 bundled OFL families chosen for the brief's character, with real kerning (Turkish and
-  other Latin-extended letters checked in every weight); every changed letter is measured against the font's own
-  alphabet, so a word that would be misread ("an e that lost its eye", a dot that reads as a hyphen) is caught before
-  review. Lockups aligned to cap height, OKLCH tonal palettes with WCAG pairs.
-- **Looks before it talks.** Directions start as cheap sketches that a critical review must pick before any are
-  developed. Every concept is rendered to a review sheet (16–256 px, light, dark, black, white) and a context sheet (a
-  browser tab, a home screen next to the competitors, a site header, an avatar). An independent critic records a blind
-  first impression before it reads the rationale, and only concepts that pass its gate (brand fit, distinctiveness and
-  memorability ≥ 6 and "a senior designer would present it") reach you. A first look on the edge of the bar is decided
-  by the majority of three cold looks from fresh critics, and the designer can never approve its own work.
-- **Complete asset kit.** Logo variants, favicons, PWA + maskable icons, iOS (incl. dark/tinted and Icon Composer
-  layers), macOS `.icns`, Android adaptive + themed icons, Windows ICO + MSIX assets, Open Graph and social images,
-  marketing templates, design tokens and brand guidelines, with a manifest of sizes and hashes. You get what the
-  product needs: a web app gets web, social, tokens and guidelines; app and desktop icon sets are one question away.
-- **Brand Preview Gallery.** One offline HTML file comparing concepts on realistic surfaces (browser tab, home screen,
-  Dock, store listing, link preview, business card). Pin notes on the mark itself, compare two concepts side by side,
-  mark a preferred one, and send the decisions straight back to Claude (`gallery --serve`, local only) or paste them.
-- **Monorepo-aware.** Finds the sites, apps and packages of a monorepo, registers the user-facing ones as brands of a
-  family, and researches each separately — including existing icon sets and design-system docs.
-- **Families and locked brands.** Multi-product families share a system; existing identities can be **locked** so they
-  are reproduced byte-for-byte and never redesigned.
-- **Fits your stack.** `install-assets --framework auto` puts the files where Next.js (App Router metadata files),
-  Laravel, Flutter (Android, iOS and web) or a static site expects them — a dry run first, backups before anything is
-  replaced.
-- **Not tied to one AI.** Claude Code is home, but the toolchain is a plain CLI: `agent-setup` gives VS Code with
-  Copilot, Codex and Cursor the same workflow, with the researcher, designer and critic as sub-agents on the models
-  you choose (see [docs/INTEGRATIONS.md](docs/INTEGRATIONS.md)).
-
-## Measured quality
-
-Markilo is benchmarked end to end with nobody steering, on nine realistic products (an independent SaaS, a mobile
-app, a multi-product family, a developer tool, an online shop, a B2B platform, a learning app, a feature-flag service
-and a team secrets manager). Six of them were **holdouts** for a while: written before the changes they test and
-never used for tuning until those results were in. The plugin researches, writes the brief, designs, critiques, and
-finalizes only what passes its own gate. Blind judges then rate every critiqued concept from the brief and the images
-alone. Every run, image and score is in [bench/](bench).
-
-| Version      | Runs                                        | Passed the gate          | Opus 5.5 judge would present (of concepts it judged) | Time per run | Cost per run (API equivalent) |
-| ------------ | ------------------------------------------- | ------------------------ | ---------------------------------------------------- | ------------ | ----------------------------- |
-| 0.4.0-rc.1   | 3                                           | 0                        | —                                                    | 11–21 min    | $2.61–3.67                    |
-| 0.4.0-rc.2   | 3                                           | 1                        | —                                                    | 8–15 min     | $1.60–2.68                    |
-| 0.4.0-rc.3   | 3                                           | 2                        | 0 of 2 (the two that passed)                         | 14–24 min    | $3.15–4.37                    |
-| 0.4.0-rc.5   | 9 (6 on holdouts)                           | 2                        | 0 of 12                                              | 10–19 min    | $2.29–4.41                    |
-| 0.4.0-rc.7   | 11 (2 new holdouts)                         | 0                        | 2 of 5 — both rejected by the gate                   | 6–16 min     | $1.66–3.30                    |
-| 0.4.0-rc.8   | 9                                           | 0                        | 0 of 3 — the recalibrated gate agreed on all 3       | 8–18 min     | $1.91–3.48                    |
-| 0.4.0 (rc.9) | 9                                           | 2                        | Portwise 2 of 4 looks, Dozeline 0 of 4               | 14–28 min    | $3.00–7.13                    |
-| 0.5.0-rc.1   | 9 (4 on new holdouts)                       | 2                        | 2 of 29, the two that passed (3/3 and 2/3 looks)     | 14–36 min    | $3.37–5.86                    |
-| **1.0.0**    | **14 (5 sealed holdouts, 9 on real tools)** | **2, both false passes** | **0 of 43 (Opus + Sonnet panel)**                    | 21–35 min    | $6.85–13.87                   |
-
-What this shows, plainly:
-
-- **1.0 did not make unattended work better, and says so.** On five briefs sealed before any 1.0 change and on three
-  real developer tools, an independent panel (Opus and Sonnet, blind) would present none of the 1.0 concepts; it would
-  not have presented the 0.5.0-rc.1 ones either. The two 1.0 gate passes were edge passes the panel rejected; the
-  gallery now marks such passes "at the edge". What 1.0 did improve is everything around the design: safety, the
-  install transaction, letter checks, typefaces, the critique process and the evidence. See
-  [bench/results/1.0.0](bench/results/1.0.0).
+<p align="center">
+  <img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/screens/gallery.png" alt="The Markilo Gallery showing three proposals, A Hallmark, B Inktrap and C Seal, side by side" width="100%">
+</p>
+<p align="center"><sub>The Gallery with Markilo's own round of three proposals (from <a href="https://github.com/cihantopac/markilo/tree/main/branding">branding/</a>; the chosen one, A, already shows its finished icon).</sub></p>
 
-- **Unattended, Markilo still rarely reaches its own gate** — 6 of the 47 validation runs since rc.5, 2 of 12 on
-  briefs that were still untouched holdouts when they ran.
-- **For the first time, what passes is what a strict judge would present.** In 0.5.0-rc.1 the gate and the Opus judge
-  (the majority of three looks per concept) agreed on all 29 critiqued concepts, and both identities that passed —
-  Quillnest, and Pinehold on a brief written for this round — were presentable to the judge. In 0.4.0 the two that
-  passed split it 2 of 4 and 0 of 4.
-- **Words beat invented symbols.** Both passes were wordmark-led: the name drawn from a real typeface with one letter
-  changed, the icon cut from it (2 of 13 judged presentable; invented symbols 0 of 16). Symbols assembled from
-  rectangles, circles and polygons keep reading as UI icons or objects — a power button, a wallet, an eye.
-- **Reworking a failed idea does not work, so Markilo stopped doing it.** Of 34 concepts sent back after a failed gate
-  in earlier rounds, one passed later, after three rounds. The gate now says `fix` once or `reject`; in rc.1 one
-  concept got a fix round, and the budget went to more directions instead (43, up from 27).
-- **How the gate got here was measured.** The critic's scores given after reading the rationale barely tracked the
-  strict judge (Spearman 0.20 on 32 concepts), a cold look did much better (0.51); the gate rests on the cold look.
-  Across all runs the gate and the judges agreed on 77 of 86 critiqued concepts. When nothing passes, Markilo says
-  so and nothing is finalized or exported.
-- **What moved the needle was measured, not guessed.** Hairline gaps came from Markilo's own guide; clichés reached
-  development because the sketch review picked the best of weak boards; the name's initial with a cut became the
-  default move once objects were ruled out; same-colour layers left seams. Each was found in the runs and fixed in
-  the plugin. The model was not the bottleneck (Opus 5.5 as designer did no better on the SaaS brief).
+## Three proposals → Compare → Select or request changes → Finalize → Export
 
-Markilo is built to work with you: you pick the sketch to develop, react in the gallery and decide. Unattended runs
-measure the floor. The two that passed in 0.5.0-rc.1, as exported (fictional products, no human edits):
+Markilo reads your product, writes a short brief and draws **three genuinely different identities**. In the Gallery you
+see each one on light and dark, as an app icon and at 16 px, with its idea, its strengths and its honest weaknesses.
+Nothing is finalized without your choice.
 
 <table>
   <tr>
-    <td><img src="bench/results/0.5.0-rc.1/edu-app/quillnest/overview.png" alt="Quillnest logo files made by Markilo unattended"></td>
-    <td><img src="bench/results/0.5.0-rc.1/secrets-team-r2/pinehold/overview.png" alt="Pinehold logo files made by Markilo unattended"></td>
+    <td width="34%" valign="top"><img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/screens/gallery-request.png" alt="A proposal with Request changes chosen: chips for simpler and better at small sizes, a note, and a pin on the mark"></td>
+    <td valign="top"><img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/screens/gallery-compare.png" alt="Two proposals compared side by side, each with a Prefer this button"></td>
   </tr>
   <tr>
-    <td><sub>Quillnest (writing coach for teenagers): passed the gate; Opus judge "present" in 3 of 3 looks, 6.3 — "a
-    smiling face built from a q, with the orange dot as a nose".</sub></td>
-    <td><sub>Pinehold (secrets manager for small teams, a fresh holdout): passed the gate; Opus judge "present" in 2 of 3
-    looks, 6.3 — "an arch or doorway sheltering an orange seed".</sub></td>
+    <td><sub><b>Request changes</b> in your words: pick what to change, add a note, or pin a spot on the mark.</sub></td>
+    <td><sub><b>Compare</b> any two side by side and say which you prefer and why. Your decisions go straight back to Claude.</sub></td>
   </tr>
 </table>
 
-## Install
-
-Requirements: [Claude Code](https://code.claude.com) and Node.js ≥ 20.10. Nothing else — the toolchain ships prebuilt
-(pure JavaScript + WebAssembly) and works on macOS, Linux and Windows.
-
-In Claude Code:
-
-```text
-/plugin marketplace add cihantopac/markilo
-/plugin install markilo@markilo
-```
-
-Or from a terminal:
-
-```bash
-claude plugin marketplace add cihantopac/markilo
-```
-
-```bash
-claude plugin install markilo@markilo
-```
-
-To update later: `claude plugin marketplace update markilo`, then `claude plugin update markilo@markilo` and restart
-Claude Code.
-
-Without Claude Code — for other coding agents ([docs/INTEGRATIONS.md](docs/INTEGRATIONS.md)) and scripts — the same
-toolchain is on npm:
-
-```bash
-npm install -g markilo
-```
-
-Then `markilo help`; to run without a global install, use `npx markilo <command>`.
-
-### `/markilo` and `/markilo:design`
-
-After installing, both `/markilo` and `/markilo:design` start the studio (verified with Claude Code 2.1.308). If another
-command already uses the name `/markilo` on your machine, Markilo can install a tiny personal skill that forwards to the
-plugin. It never touches files it did not create:
-
-```bash
-node ~/.claude/plugins/marketplaces/markilo/plugin/dist/markilo.mjs shortcut install --yes
-```
-
-(`shortcut status` and `shortcut uninstall --yes` are available too.)
+## Made with Markilo
 
-## Use
+Real products, each from one round of three proposals with its owner choosing. These are the files Markilo exported,
+unedited.
 
-| Command                                        | What it does                                                                                                  |
-| ---------------------------------------------- | ------------------------------------------------------------------------------------------------------------- |
-| `/markilo:design <url, repo, name or task>`    | The full studio: discover → research → brief → explore → design → verify → review → refine → export → deliver |
-| `/markilo:research <url or repo>`              | Brand Intelligence Profile and brief only                                                                     |
-| `/markilo:concepts <brand>`                    | Creative directions and verified SVG concepts                                                                 |
-| `/markilo:originality <concept, file or name>` | Similarity analysis, name check and trademark pre-screening plan                                              |
-| `/markilo:refine <brand> <concept> <feedback>` | Iterate with verification, then finalize                                                                      |
-| `/markilo:export <brand>`                      | Generate the asset kit (optionally install web assets into your project)                                      |
-| `/markilo:gallery`                             | Build the Brand Preview Gallery                                                                               |
-| `/markilo:audit <logo or brand>`               | Evaluate an existing identity: keep, refresh or redesign                                                      |
+<table>
+  <tr>
+    <td width="33%"><a href="https://github.com/cihantopac/relivaro"><img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/made-with/relivaro-og.png" alt="Relivaro identity: a cairn of stones and the word Relivaro"></a></td>
+    <td width="33%"><a href="https://github.com/cihantopac/searvex"><img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/made-with/searvex-og.png" alt="Searvex identity: a broad-nib V and the word searvex"></a></td>
+    <td width="33%"><a href="https://github.com/cihantopac/jourilo"><img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/made-with/jourilo-og.png" alt="Jourilo identity: a book serif with a hanging J"></a></td>
+  </tr>
+  <tr>
+    <td><sub><b>Relivaro</b> — release engineering. ≈ $15, seven runs while the 1.1 process was being corrected.</sub></td>
+    <td><sub><b>Searvex</b> — search intelligence. $11.27: one round, one development, one direct change.</sub></td>
+    <td><sub><b>Jourilo</b> — product experience. $5.29: one round, finished as chosen.</sub></td>
+  </tr>
+</table>
 
-Requests can be written in any language; files and reports are produced in English.
+Markilo's own identity came out of the same process: $13.68 for one round and three icon options.
 
-### A typical session
+## The kit you get
 
-```text
-you  › /markilo:design ./ — we're launching this as a product, it needs an identity
-     › (Markilo scans the repo: package.json, README, existing favicon, CSS colors, fonts)
-     › Discover: no real identity yet → mode "new". May I research competitors on the web? (y)
-     › Research: 8 competitors, category codes (green + shields everywhere), 3 gaps recorded
-     › Brief: 3 questions, each with a suggested answer
-     › Ideas: 16 visual hooks; the critic strikes 11 from the words ("a gauge", "a return key"), keeps 5
-     › Explore: 3 directions from the kept ideas — "Pulse" monogram, "Lighthouse" pictogram, "Signal" abstract
-     › Sketch: 4 quick constructions per direction; you pick in the gallery, with the critic's reading beside each
-     › Design: concept-designer agents develop the picks; each iterates on its review and context sheets
-     › Verify: "Lighthouse" 71% similar to a competitor reference → reworked from the idea up
-     › Critique: blind first look ("reads as: a pin, a drop") then scores; only "Pulse" passes the gate
-     › Review: gallery at http://127.0.0.1:…/ — pin a note on the stem, compare, "Send to your agent"
-     › Deliverables: a web app → logos, web, social, tokens, guidelines (app icons are one question away)
-     › Finalize "Pulse" → Export: 80 files → Next.js detected: 10 files for app/ and public/ (dry run, 0 overwrites)
-```
+<p align="center">
+  <img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/kit-overview.png" alt="Every Markilo logo file: primary, secondary, compact, symbol and wordmark in color, reversed, black and white" width="100%">
+</p>
 
-### Monorepos and product families
+Logos in color, reversed, black and white (SVG + PNG) · favicons, PWA and maskable icons · iOS, macOS, Android and
+Windows app icons · Open Graph and social images · design tokens · brand guidelines · a manifest with sizes and
+checksums. Each kit opens with a preview page that shows every file on the background it is made for:
 
-```text
-/markilo:research This is our product family monorepo at ./ — sites in products/, apps in apps/. The Ledger app's existing identity is locked.
-```
-
-Markilo runs `research monorepo`, shows you the packages it considers brands, and after you confirm registers them
-(with their folders) and scans each one. Work then proceeds family-first: shared system, master brand, sites one by one,
-and many apps as one coherent app-icon system. Exports can be copied into any package with
-`install-assets --project <folder>` (dry run first).
+<p align="center">
+  <img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/screens/kit-preview.png" alt="The asset kit preview page for Markilo: the primary logo on paper and on dark, the palette and a Start here table" width="100%">
+</p>
 
-## What you get
+The Gallery, the kit preview and the guidelines work offline from a single file, in **English or Turkish**, in
+**light, dark or system** theme — on a phone too.
 
-```text
-brand/
-├── markilo.json                    # workspace: brands, family
-├── brands/<slug>/
-│   ├── brief.json · profile.json · answers.json · decisions.md
-│   ├── research/references/        # competitor marks for comparison (git-ignored)
-│   ├── concepts/<id>/              # concept.json, symbol.svg, wordmark.svg, review/*.png
-│   ├── reports/                    # originality-<id>.md|json, quality-<id>.md|json
-│   ├── final/                      # approved masters + brand.json
-│   └── exports/
-│       ├── logos/                  # primary, secondary, compact, symbol, wordmark, appmark × color/reversed/black/white/ink (SVG + PNG)
-│       ├── web/                    # favicon.svg (dark-mode aware), favicon.ico, apple-touch-icon, icon-*/maskable-*, site.webmanifest, head-snippet.html
-│       ├── ios/                    # AppIcon.appiconset (default, dark, tinted), icon-composer/ layers
-│       ├── macos/                  # AppIcon.icns, .iconset, full-bleed 1024 for macOS 26+
-│       ├── android/                # mipmap-* legacy + adaptive foreground/background/monochrome, anydpi-v26 XML, Play icon
-│       ├── windows/                # app.ico, MSIX Assets (scales, target sizes, unplated variants)
-│       ├── social/                 # og-image (+dark), GitHub social preview, X header, LinkedIn cover, avatar
-│       ├── marketing/              # watermarks, email signature, title slide, business card
-│       ├── tokens/                 # tokens.json (W3C DTCG), tokens.css, palette.json
-│       ├── guidelines/             # brand-guidelines.md, index.html
-│       └── asset-manifest.json     # every file: purpose, size, sha256, source master
-└── gallery/index.html              # Brand Preview Gallery
-```
+<p align="center">
+  <img src="https://raw.githubusercontent.com/cihantopac/markilo/markilo--v1.1.2/assets/screens/gallery-mobile-tr-dark.png" alt="The Gallery on a phone, in Turkish and dark theme" width="300">
+</p>
 
-Platform sizes and safe zones are kept in [`src/export/specs.ts`](src/export/specs.ts) with the official source for
-each requirement (Apple HIG and Xcode docs, Android Developers, Microsoft Learn, W3C/MDN/web.dev, OGP, GitHub, LinkedIn).
-See [docs/OUTPUTS.md](docs/OUTPUTS.md).
+## Install
 
-## How it works
+In Claude Code:
 
 ```text
-Claude Code  ──  skills/ (process: research, strategy, design, critique)  ──  agents/ (researcher, designer, critic)
-                       │  node ${CLAUDE_PLUGIN_ROOT}/dist/markilo.mjs …
-                       ▼
-Toolchain (src/ → plugin/dist/): XML/SVG lint & optimize · resvg-wasm rendering · shape features & perceptual hashes ·
-similarity & genericity · quality scorecard · fontkit wordmarks · lockups · OKLCH palettes · research scanners ·
-originality & quality reports · exporters (ICO/ICNS encoders) · gallery · guidelines
-```
-
-The rule that keeps it honest: **anything measurable comes from code; anything judged is labeled as judgment.**
-Read [docs/ARCHITECTURE.md](docs/ARCHITECTURE.md) and [docs/ORIGINALITY.md](docs/ORIGINALITY.md).
-
-## The toolchain on its own
-
-The same CLI runs without Claude Code — useful in CI or scripts:
-
-```bash
-node plugin/dist/markilo.mjs help
-```
-
-```bash
-node plugin/dist/markilo.mjs compare my-logo.svg competitor.png
+/plugin marketplace add cihantopac/markilo
+/plugin install markilo@markilo
 ```
 
-Examples: `wordmark --text "Acme" --font sora --weight 600 -o acme.svg`, `score logo.svg`, `palette "#f45d2c"`,
-`render logo.svg -o sheet.png --sheet`, `name check Acme --against Acmo,Akme`. Every command prints JSON.
-
-## Privacy and permissions
-
-- Everything runs locally, with no accounts, servers or paid APIs. The toolchain reaches the network only in the
-  research and font commands (`research site`, `research competitors`, `research app`, `fonts search/add`), which require
-  `--allow-network`; the skills ask you first, and before web searches that would reveal unreleased names. Site
-  research honours `robots.txt`.
-- The repository scanner never opens secret-like files (`.env*`, keys, credentials), and profiles store project-relative
-  paths, never your user name or folder layout.
-- Markilo writes only inside the brand workspace (`./brand`). Copying assets into your project is a dry run until you
-  approve `--apply`; existing files are never overwritten without `--force` and are backed up first.
-- `gallery --serve` listens on `127.0.0.1` only, with a random token, and serves only the gallery and approved artwork.
-- Raw research captures and competitor references are git-ignored inside the workspace.
-
-Details: [docs/PRIVACY.md](docs/PRIVACY.md) · [SECURITY.md](SECURITY.md).
-
-## Limitations
+Then: `/markilo <your repo, URL or product name>`.
 
-- **Not legal advice.** Similarity analysis and registry pre-screening do not clear a trademark. No results ≠ unique.
-  Use a trademark attorney before adopting a brand.
-- Automated similarity measures form, composition, structure and color. **Conceptual** similarity (the same idea drawn
-  differently) is judged by the critic agent and by you.
-- Reference comparison covers only the marks collected for the brand; collecting good references matters.
-- Raster images (PNG/JPEG) are analyzed for similarity; vector tracing of raster logos is not included.
-- Icon Composer (`.icon`) files are not written directly; Markilo exports the layer sources to import.
-- X (Twitter) image sizes could not be verified against x.com documentation (blocked to automated readers).
-- Design quality still depends on the model and on your direction: unattended runs rarely pass the critic gate, and a
-  strict independent judge accepts only a few of the marks they develop (see [Measured quality](#measured-quality)).
-  On real developer tools with coined names, twelve unattended runs produced no identity an independent panel would
-  present ([bench/results/1.0.0/real-projects.md](bench/results/1.0.0/real-projects.md)). Markilo raises the floor and
-  refuses to hide weak work; taste is a conversation, and the gallery is where you steer it.
-
-## Develop
+Without Claude Code — for other coding agents and scripts — the same toolchain is on npm (Node.js ≥ 20.10, macOS,
+Linux and Windows):
 
 ```bash
-npm ci
-```
-
-```bash
-npm run ci
+npm install -g markilo
 ```
 
-`npm run ci` runs Biome, TypeScript, Vitest (unit, integration and bundled-CLI tests), the build and a check that the
-committed `plugin/dist/` matches the sources. `node scripts/smoke.mjs` takes an empty folder to installed assets with
-the bundle alone (CI also runs it on the copy `claude plugin install` produces). `node scripts/bench.mjs` runs the
-end-to-end benchmark ([bench/](bench)). Plugin evals live in [`plugin/evals/`](plugin/evals) (`claude plugin eval plugin`).
-See [CONTRIBUTING.md](CONTRIBUTING.md) and [docs/MAINTAINING.md](docs/MAINTAINING.md).
+## What it costs, and what it has shown
+
+- **The toolchain is free.** It runs locally — rendering, measuring, originality checks, export — with no account,
+  server or paid service.
+- **The design work is done by the AI model and uses your Claude Code usage.** In validation, one round of three
+  proposals cost $5–10 and a finished identity $5–15 (API-equivalent, measured by Claude Code; on a subscription it
+  counts against your plan instead). `markilo cost` tracks it against a budget you set (default $12) and stops paid
+  work when it is reached.
+- **Steered, it works; unattended, it does not yet.** With their owner choosing, all three real products above reached
+  an identity the owner approved — three products and one owner's judgment, not a benchmark. Unattended, 1.0 produced
+  no identity a blind judge panel would present. The full record, with every run and image:
+  [benchmarks](https://github.com/cihantopac/markilo/blob/main/docs/BENCHMARKS.md) ·
+  [1.1 validation log](https://github.com/cihantopac/markilo/blob/main/docs/RELEASE-1.1.md).
+- **Not legal advice.** The originality checks compare against the marks Markilo collected; they cannot prove a mark is
+  unique or registrable. No trademark registry search is done for you.
+
+## Learn more
+
+[Usage and commands](https://github.com/cihantopac/markilo/blob/main/docs/USAGE.md) ·
+[Asset outputs](https://github.com/cihantopac/markilo/blob/main/docs/OUTPUTS.md) ·
+[Originality](https://github.com/cihantopac/markilo/blob/main/docs/ORIGINALITY.md) ·
+[Architecture](https://github.com/cihantopac/markilo/blob/main/docs/ARCHITECTURE.md) ·
+[Other coding agents](https://github.com/cihantopac/markilo/blob/main/docs/INTEGRATIONS.md) ·
+[Privacy](https://github.com/cihantopac/markilo/blob/main/docs/PRIVACY.md) ·
+[Security](https://github.com/cihantopac/markilo/blob/main/SECURITY.md) ·
+[Changelog](https://github.com/cihantopac/markilo/blob/main/CHANGELOG.md) ·
+[Contributing](https://github.com/cihantopac/markilo/blob/main/CONTRIBUTING.md)
 
 ## License
 
-[Apache-2.0](LICENSE). Bundled fonts are SIL OFL 1.1; see [THIRD_PARTY_NOTICES.md](THIRD_PARTY_NOTICES.md).
+[Apache-2.0](https://github.com/cihantopac/markilo/blob/main/LICENSE). Bundled fonts are SIL OFL 1.1; see
+[THIRD_PARTY_NOTICES.md](https://github.com/cihantopac/markilo/blob/main/THIRD_PARTY_NOTICES.md).

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "markilo",
-  "version": "1.0.0",
-  "description": "AI-powered brand identity studio for Claude Code: research, originality analysis, SVG identity design, verification and complete asset export.",
+  "version": "1.1.2",
+  "description": "Brand identity studio for Claude Code: three original SVG identities to compare and choose from, originality checks, and a complete asset kit (logos, app icons, favicons, social images, tokens, guidelines).",
   "license": "Apache-2.0",
   "type": "module",
   "author": "Cihan Topaç (https://github.com/cihantopac)",
@@ -22,7 +22,9 @@
     "favicon",
     "app-icon",
     "design-system",
-    "originality"
+    "originality",
+    "brand-guidelines",
+    "design-tokens"
   ],
   "engines": {
     "node": ">=20.10"

package/plugin/.claude-plugin/plugin.json

@@ -1,8 +1,8 @@
 {
   "name": "markilo",
   "displayName": "Markilo",
-  "version": "1.0.0",
-  "description": "AI-powered brand identity studio: brand research, originality and similarity analysis, original SVG identity design with visual verification, and complete asset export (web, iOS, Android, macOS, Windows, social).",
+  "version": "1.1.2",
+  "description": "Brand identity studio: researches your product, draws three original SVG identities for you to compare and choose from, checks originality, and exports a complete asset kit (web, iOS, Android, macOS, Windows, social, tokens, guidelines).",
   "author": {
     "name": "Cihan Topa\u00e7",
     "url": "https://github.com/cihantopac"