What makes Buck Mason Stylist notable?

1.3K downloads. Trust evidence available

How should Buck Mason Stylist be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Buck Mason Stylist

Personal shopping skill for Buck Mason. Stock-checks (online + nearby store), wardrobe gap analysis, season- and event-aware outfit suggestions, AI try-on lo... Skill: Buck Mason Stylist Owner: nickmerwin Summary: Personal shopping skill for Buck Mason. Stock-checks (online + nearby store), wardrobe gap analysis, season- and event-aware outfit suggestions, AI try-on lo... Tags: latest:0.7.2 Version history: v0.7.2 | 2026-05-19T16:51:28.017Z | auto Version 0.7.2 - Adds support and documentation for the agent-driven MPP checkout path via @stripe/link-cli. - Updates required ag

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.3k

Updated

Oct 10, 2026

Version

0.7.2

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.3K downloads reported by the source. Last updated 10/10/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 10, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 10, 2026
Adoption signal: 1.3K downloadsadoption · observed Oct 10, 2026
Latest release: 0.7.2release · observed May 19, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17dybwreermy66eaezsvfsgmn85w1pw:buck-mason-stylist-skill

Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-nickmerwin-buck-mason-stylist-skill/snapshot"

Documentation

CLAWHUB

160,000 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: buck-mason-stylist
description: Personal shopping skill for Buck Mason. Stock-checks (online + nearby store), wardrobe gap analysis, season- and event-aware outfit suggestions, AI try-on lookbooks, and one-shot MPP checkout via link-cli. Customer brings sizes once; the agent reuses them across requests.
version: 0.7.2
license: MIT
authors:
  - Buck Mason / Pima
runtime: any
compatibility:
  mcp_servers: [pima-mcp]
  binaries: [curl, jq]
metadata:
  openclaw:
    requires:
      binaries: [curl, jq]
      python: [python-pptx, Pillow]
      optional_env: [OPENAI_API_KEY]
      optional_binaries: [magick]
      optional_clis: ["@stripe/link-cli"]
    env_details:
      OPENAI_API_KEY:
        required: false
        conditional: "Default for any unqualified lookbook request: Premium-tier gpt-image-2 virtual try-on imagery, with Editorial/Minimum used only when Premium prerequisites are missing, fail, or the customer explicitly asks for no AI. Stock checks, recommend, MPP checkout, and order tracking are all unaffected by absence."
        purpose: "OpenAI /v1/images/edits with model gpt-image-2 for AI try-on imagery."
        format: "OpenAI API secret key, prefix sk-..."
        obtain_url: "https://platform.openai.com/api-keys"
        notes: "The OpenAI organization must be verified for gpt-image-2 access. Unverified orgs get 403 from /v1/images/edits; the skill surfaces that as an actionable error and offers Editorial tier rather than silently downgrading. See https://help.openai.com/en/articles/10910291."
        how_to_set: "export OPENAI_API_KEY=sk-..."
    optional_cli_details:
      "@stripe/link-cli":
        purpose: "Required only for the fully-agent-driven MPP checkout path (workflow #4 step 3). Mints a one-time Stripe Shared Payment Token from the customer's Link wallet."
        install: "npm i -g @stripe/link-cli"
        publisher: "Stripe (verified npm publisher)"
        source: "https://github.com/stripe/link-cli"
        npm: "https://www.npmjs.com/package/@stripe/link-cli"
        notes: "Install only from the @stripe scope. Pin to a reviewed version in production. The skill never bundles or vendors this CLI."
    categories: [commerce, image-generation, lookbook]
    tags: [buck-mason, pima, stylist, shopping, stock, mcp]
---

# Buck Mason personal stylist

You are acting as a personal shopper for Buck Mason. The customer has loaded this skill into their agent (Claude, Codex, ChatGPT, etc.) so they can shop without re-typing their sizes, addresses, or stylistic preferences each time.

> **What's loaded by agents at runtime, vs repo-only.** Follow refs from this `SKILL.md` only. Files like `README.md`, `PUBLISHING.md`, `SECURITY.md`, and `CLAUDE.md` exist for repo readers, ClawHub reviewers, and skill editors — agents should not load them at runtime. Anything an agent needs is either in the frontmatter, in `references/`, in `templates/`, in `scripts/`, or in `examples/`.

## Environment

**Premium try-on uses one opt

README.md

# Buck Mason Stylist Skill

A personal-shopping skill for [Buck Mason](https://www.buckmason.com), built for Claude Code, Codex, ChatGPT custom GPTs, and any agent that loads `SKILL.md`–style skills. Talks to the pima.io MCP at `pima.io/mcp/buckmason/*`.

## What it does

- **Stock check** — "do they have the [item] in my size, online and at the Abbot Kinney store?" — returns bucketed live counts (`In stock` / `Low stock (N left)` / `Out of stock`) per location.
- **Wardrobe gap analysis** — "what am I missing for a Sonoma wedding in May?" — diffs your owned items against a season + climate + dress-code-aware capsule recommendation, with a one-sentence "why" per pick.
- **AI try-on lookbooks** — by default, "lookbook" means gpt-image-2 virtual try-on images of you wearing the recommended outfits, assembled into a hosted HTML/HTML-cart lookbook with partner voting enabled.
- **One-shot MPP checkout** — `POST /mcp/buckmason/checkout` speaks the [Merchant Payments Protocol](https://mpp.dev): HTTP 402 + Stripe Shared Payment Token via [`stripe/link-cli`](https://github.com/stripe/link-cli), push-approved by the customer in the Link app. See `references/mpp.md`.

## Required setup

### `OPENAI_API_KEY` (for the default AI try-on lookbook workflow)

```bash
export OPENAI_API_KEY=sk-...
```

The skill posts to `https://api.openai.com/v1/images/edits` with `model: "gpt-image-2"` to generate virtual try-on images. **The OpenAI organization tied to the key must be verified for `gpt-image-2`** (see <https://help.openai.com/en/articles/10910291>). Get a key at <https://platform.openai.com/api-keys>.

The other workflows — stock check, recommend, checkout, order tracking — do **not** require an OpenAI key. They only call the pima.io MCP. If a lookbook request is missing the key or usable reference photos, the skill should say what is missing and offer the Editorial tier instead of silently downgrading.

### Voting on deployed lookbooks

Cloudflare Pages deploys include the voting mechanism by default. Create one KV namespace per Cloudflare account, then save the id in `profile.md` as `lookbook_votes_kv_id:` or export it as `LOOKBOOK_VOTES_KV_ID`.

```bash
CLOUDFLARE_ACCOUNT_ID=<account-id> wrangler kv namespace create LOOKBOOK_VOTES
```

### Profile

One profile file in your agent's workspace (copy from `templates/profile.example.md`):

```bash
cp templates/profile.example.md ~/agent-workspace/profile.md
$EDITOR ~/agent-workspace/profile.md   # fill in sizes, home zip, build, face, reference photos
```

That's it. No Buck Mason credentials. No Pima account.

Optional but useful:
- `wardrobe.md` — owned items (enables gap analysis)
- `events.md` — upcoming travel/events (enables event-aware suggestions)
- `magick`, `python-pptx`, `Pillow` — only needed for the PPT/HTML lookbook output (see `references/output-formats.md`)
- [`stripe/link-cli`](https://github.com/stripe/link-cli) — required for the MPP fully-agent-driven checkout path

For MPP checkout, `profile

_meta.json

{
  "ownerId": "kn735q4jn33x2q1w5fpe46kqwd85wzq0",
  "slug": "buck-mason-stylist-skill",
  "version": "0.7.2",
  "publishedAt": 1779209488017
}

references/acceptance-checklist.md

# Lookbook acceptance checklist

The single source of truth for "is this lookbook ready to share with the customer." Every `html` / `html-cart` / `ppt` lookbook clears these gates **before** the agent emits the URL or filename. The checklist is operationalized as `scripts/validate-lookbook.py` — run it; if it exits non-zero, fix the failure rather than narrating around it.

## Local checks (run against the deploy directory before upload)

| # | Check | Pass criterion |
|---|---|---|
| L1 | `index.html` exists | File present, ≥ 4 KB |
| L2 | All product links absolute + on-brand | Every `<a href>` matching `/products/` resolves to `https://www.buckmason.com/products/<slug>` |
| L3 | Prices present per piece | At least one `$\d+` token within each `.piece` block |
| L4 | Stock lines present per piece | Each `.piece` carries an `In stock` / `Low (N)` / `Out of stock` substring (the bucket strings from `/mcp/buckmason/stock`) |
| L5 | AI try-on disclosure present (premium tier only) | Page contains "AI-generated" or "AI try-on" text, in cover or footer — required when any look hero came from gpt-image-2 |
| L6 | OG meta tags present + absolute | `og:url`, `og:image`, `twitter:image` each resolve to URLs starting with `https://`, **not** containing `{{ABSOLUTE_PAGE_URL}}` / `{{ABSOLUTE_OG_IMAGE_URL}}` placeholders |
| L7 | `og.jpg` present + correctly sized (multi-file deploys) | File exists in deploy dir, ~1200×630, `.jpg` mimetype, < 500 KB |
| L8 | Per-piece checkboxes carry the full data set (`html-cart` only) | Every `.piece input[type="checkbox"]` has `data-name`, `data-size`, `data-sku`, `data-qty`, `data-price-cents` |
| L9 | Subtotals and look totals match piece prices | Per-look total equals the sum of piece prices in that look (rendered via JS only — verify the data attrs sum cleanly) |
| L10 | No broken inline references | No `data-fullsize=""`, no `<img src="">`, no `href="#"` placeholders |

## Deployed checks (run after `wrangler pages deploy` against the live URL)

| # | Check | Pass criterion |
|---|---|---|
| D1 | Page returns HTTP 200 | `curl -sI <page-url>` first line is `HTTP/2 200` (or `HTTP/1.1 200 OK`) |
| D2 | `og.jpg` returns HTTP 200 + `image/jpeg` | `curl -sI <og-url>` shows `200` + `content-type: image/jpeg` |
| D3 | Each `look<N>.jpg` returns 200 (multi-file deploys) | One per look section; if any 404s, redeploy with the missing asset |
| D4 | Meta tags survived the deploy | The page body still contains all 13 OG/Twitter tags (no template substitution dropped them) |
| D5 | Resolved URLs in OG metadata are reachable | The `og:url` and `og:image` URLs in the served HTML themselves return 200 (catches mismatched-alias bugs where the page deployed but the OG URL points at a stale/wrong subdomain) |
| D6 | Unfurl preview works | Hit `https://www.opengraph.xyz/url/<encoded-url>` in a browser, or scrape with a `User-Agent: facebookexternalhit/1.1` and confirm the meta tags are visible |

## Warnings (non-blocking, surface to th

references/brand-style.md

# Buck Mason brand style guide

A snapshot of the live storefront's visual language, extracted from `buckmason.com` directly (homepage + `/collections/curved-hem-tees` + `/products/white-slub-curved-hem-tee`) on **2026-05-05**. Use this when rendering any branded surface inside the skill — the `html-cart` lookbook in particular — so the output reads as "by Buck Mason," not "AI-generated for Buck Mason."

For higher-stakes artifacts the agent can optionally re-extract live (script in the appendix). Treat this file as a fast default; trust the live site if they diverge.

## At a glance

- **One typeface family does almost everything**: Adobe **Acumin Pro** (regular + condensed). Body text is regular Acumin Pro; every label, headline, button, and nav item is **Acumin Pro Condensed, UPPERCASE, with +2% letter-spacing**. There are no serifs anywhere on the standard catalog. Don't reach for Georgia, Canela, Söhne, or any "editorial serif" instinct — that's not how Buck Mason looks.
- **The page is white.** `#FFFFFF` body, `#333333` text. Off-white `#F3F1EF` shows up only as an accent (header announcement banner). Don't paint the whole page off-white; that reads more J.Crew than Buck Mason.
- **Imagery does the heavy lifting.** Headlines are small (~20px), tightly scaled, never bigger than the body's gallery photos. The brand voice is conveyed by hero photography and copy decks, not by display typography.
- **Sharp corners.** `border-radius: 1px` (≈ 0) on CTAs, size selectors, and product tiles. Pills only on round badges (color swatches, etc., where `border-radius: 50%`).
- **3:4 portrait product crops** (`0.75` aspect). Not 4:5, not 1:1, not 16:9.

## Colors

| Role | Value | Where it shows up |
|---|---|---|
| Page background | `#FFFFFF` (rgb 255,255,255) | `<html>`, `<body>`, `<main>` |
| Primary text | `#333333` (rgb 51,51,51) | Body, headings, prices |
| Active CTA | `#000000` on white text | "Add to Bag" once a size is selected (inferred from disabled state) |
| Disabled CTA | `#AAAAAA` background, white text | "Select Size to Add to Cart" pre-size-pick |
| Accent neutral | `#F3F1EF` (warm off-white) | Header announcement banner; subtle section panels |
| Disabled-element text | `#D9D9D9` | Sold-out size selectors |
| Border / hairline (when used) | `rgba(0,0,0,0.08)`-ish | Product tiles, form fields |

The site is otherwise **chromatic only through product imagery** — no brand reds/greens/blues. Stick to the white + black + grayscale stack and let the photos carry color.

## Typography

### Stacks

```css
/* Body */
font-family: acumin-pro, Helvetica, sans-serif;

/* Headlines, labels, nav, buttons — every cap/eyebrow/CTA */
font-family: acumin-pro-condensed, Helvetica, sans-serif;
```

Acumin Pro / Acumin Pro Condensed are licensed via Adobe Fonts. If the rendered surface is a self-contained file (`html-cart`) and we deliberately don't `<link>` external font CDNs (per `references/output-formats.md`), fall back to **`Helvetica Neue Condensed`**

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.