# BeautySpark (remote · mcp.beautyspark.me)

BeautySpark MCP server for ChatGPT, Claude and other MCP clients

- Trust score: 66/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `mcp.beautyspark.me`: 66/100 (this document), [markdown](https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp.md), [page](https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp)

## Channel facts

- Endpoint: `https://mcp.beautyspark.me/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 57/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (update_my_look).
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 43/100
  - 0% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 3641 tokens (~140/item across 26 items; 22 tools + 4 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 53/100
  - Stability observed for 16 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 99/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 96% of tool parameters carry a description.
  - Structured output schemas are declared (95% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 24 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).
  - Supports UI / widget rendering.

## Install

### How do I install the BeautySpark MCP server?

BeautySpark is a hosted endpoint at https://mcp.beautyspark.me/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http thomasfindlay-beautyspark-mcp 'https://mcp.beautyspark.me/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "thomasfindlay-beautyspark-mcp": {
      "url": "https://mcp.beautyspark.me/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "thomasfindlay-beautyspark-mcp": {
      "type": "http",
      "url": "https://mcp.beautyspark.me/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.thomasfindlay-beautyspark-mcp]
url = "https://mcp.beautyspark.me/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "thomasfindlay-beautyspark-mcp": {
      "type": "remote",
      "url": "https://mcp.beautyspark.me/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add thomasfindlay-beautyspark-mcp --url 'https://mcp.beautyspark.me/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  thomasfindlay-beautyspark-mcp:
    url: "https://mcp.beautyspark.me/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "thomasfindlay-beautyspark-mcp": {
      "Transport": "http",
      "Url": "https://mcp.beautyspark.me/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add thomasfindlay-beautyspark-mcp -t streamable-http -u 'https://mcp.beautyspark.me/mcp'
```

### Other

```json
{
  "mcpServers": {
    "thomasfindlay-beautyspark-mcp": {
      "type": "http",
      "url": "https://mcp.beautyspark.me/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 66, +1)

No change was recorded against any check on this day. Stability & Change Management went from 50 to 53. That category is still filling its 30-day observation window: 15 days of observed history at the previous scan, 16 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-19 (score 65, +1)

No change was recorded against any check on this day. Stability & Change Management went from 43 to 47. That category is still filling its 30-day observation window: 13 days of observed history at the previous scan, 14 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-16 (score 64, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-14 (score 63, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-12 (score 62, +1)

- [security] Tool “save_my_look” rewrote its description, which is the text the model reads

### 2026-11-10 (score 61, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-08 (score 60, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-06 (score 59, +1)

- [functional improvement] Stability: unverified → 0.03

## MCP tools (22)

### `search_guides` (~210 tokens)

Search BeautySpark guides

Search BeautySpark's makeup and colour-season guides for sections that answer a question, e.g. eyeliner for hooded eyes or how to find your undertone. Returns up to 5 sections with a short snippet and a link; read one in full with get_guide. Cite the returned url. When nothing matches, the result has a note saying no BeautySpark guide covers it.

Input parameters:

- `limit` (integer): Results to return, 1 to 5. Defaults to 5.
- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.
- `query` (string, required): Keywords for the makeup or colour question, e.g. "eyeliner hooded eyes".
- `season` (string): Color season, e.g. "True Winter" or "true_winter". One of the 12 seasons.
- `type` (string): Only articles of this type.

Output parameters:

- `note` (string)
- `results` (array)

### `get_guide` (~78 tokens)

Read a BeautySpark guide section

Read one section of a BeautySpark guide in full, by an id from search_guides. Also lists the ids and headings of the other sections of the same article. Cite the returned url.

Input parameters:

- `id` (string, required): A section id returned by search_guides, e.g. "en/how-to-find-your-undertone#intro".

Output parameters:

- `id` (string)
- `lang` (string)
- `metadata` (object)
- `text` (string)
- `title` (string)
- `url` (string)

### `search` (~118 tokens)

Search BeautySpark (research)

For research modes (deep research, company knowledge). In chat, use search_guides instead. Searches all public BeautySpark content: guide sections (id "<locale>/<slug>#<anchor>"), article looks (id "look:<id>") and colour-season palette pages (id "season:<season>"). Read any result with fetch.

Input parameters:

- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.
- `query` (string, required)

Output parameters:

- `results` (array)

### `fetch` (~119 tokens)

Fetch BeautySpark content (research)

For research modes (deep research, company knowledge). In chat, use get_guide instead. Returns the full text of one item from search: a guide section ("<locale>/<slug>#<anchor>"), an article look ("look:<id>") or a colour-season palette page ("season:<season>").

Input parameters:

- `id` (string, required): An id returned by search.
- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.

Output parameters:

- `id` (string)
- `metadata` (object)
- `text` (string)
- `title` (string)
- `url` (string)

### `get_color_season` (~118 tokens)

Get color season palette

BeautySpark's palette for one of the 12 color seasons: best, neutral and avoid colors with hex codes, plus the eye makeup and lips & cheeks advice from the season guide. Cite the returned URLs.

Input parameters:

- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.
- `season` (string, required): Color season, e.g. "True Winter" or "true_winter". One of the 12 seasons.

Output parameters:

- `advice` (array)
- `displayName` (string)
- `guideUrl` (string)
- `lang` (string)
- `season` (string)
- `swatches` (object)

### `get_product_picks` (~134 tokens)

Get product picks for a color season

BeautySpark's top product pick per category (blush, highlighter, eyeliner, lipstick, lip gloss, bronzer, eyeshadow palette) for a color season, each with the reason and the article it comes from.

Input parameters:

- `category` (string): Only return picks of this category.
- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.
- `season` (string, required): Color season, e.g. "True Winter" or "true_winter". One of the 12 seasons.

Output parameters:

- `picks` (array)
- `season` (string)

### `check_product_for_season` (~174 tokens)

Check a product against the BeautySpark picks

Looks a product (and optional shade) up in BeautySpark's season picks. Returns the seasons it is recommended for, or match: null when BeautySpark has not reviewed it; then do not present any verdict as BeautySpark's. Pass season to also get that season's palette for comparison.

Input parameters:

- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.
- `product` (string, required): Brand and product name, e.g. "ILIA Overglaze".
- `season` (string): Color season, e.g. "True Winter" or "true_winter". One of the 12 seasons.
- `shade` (string): Shade name, e.g. "Flambé".

Output parameters:

- `match`
- `seasonPalette` (object)

### `find_article_looks` (~180 tokens)

Find article looks

Find makeup looks shown in BeautySpark articles, filtered by color season, article type and words. With a season, looks made for that season come first; general looks (no season) follow and are labelled. Returns look ids for get_article_look and the article URL to cite.

Input parameters:

- `limit` (integer)
- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.
- `query` (string): Words to match against the look title, tags and image description, e.g. "smoky", "graphic liner".
- `season` (string): Color season, e.g. "True Winter" or "true_winter". One of the 12 seasons.
- `type` (string): Article type.

Output parameters:

- `looks` (array)

### `get_article_look` (~117 tokens)

Get article look

Get the steps and shades of one makeup look from a BeautySpark article. The id is a look id returned by find_article_looks ("/images/blog/<article-slug>/<look-name>").

Input parameters:

- `id` (string, required): A look id from find_article_looks, e.g. "/images/blog/<article-slug>/<look-name>".
- `locale` (string): The user's language as a language tag, e.g. "de" or "pt-BR". Set it from the user's language. Defaults to English.

Output parameters:

- `id` (string)
- `shades`
- `steps`
- `thumbnailUrl` (string)
- `title` (string)
- `url` (string)
- `webUrl` (string)

### `search_my_looks` (~157 tokens)

Search my looks

Search the user's saved looks in BeautySpark My Looks by words, tag, category, favourite or person. Returns look ids for get_my_look.

Input parameters:

- `category` (string): Only looks in this category (preset slug or the user's own category name).
- `cursor` (string): nextCursor from the previous page.
- `favourite` (boolean): true for favourites only.
- `limit` (integer): Looks per page (default 20).
- `person` (string): Only looks for this person (exact name, any case).
- `q` (string): Text to find in look titles, notes, product, shade and palette names. Use tag for tags.
- `tag` (string): Only looks with this tag name.

Output parameters:

- `looks` (array)
- `nextCursor` (string|null)
- `total` (integer)

### `get_my_look` (~56 tokens)

Get my look

Get one saved look in full: steps, products and shades, tags, categories, notes, photos and a link to open it in BeautySpark.

Input parameters:

- `id` (string, required): A saved look id from search_my_looks.

Output parameters:

- `categories` (array)
- `category`
- `createdAt` (string)
- `difficultyLevel` (string|null)
- `estimatedTotalTime`
- `id` (string)
- `inspirationUrl` (string|null)
- `intensityLevel` (string|null)
- `isFavourite` (boolean)
- `notes` (string|null)
- `occasions` (array)
- `personId` (string|null)
- `photos` (array)
- `products` (array)
- `sourceJobId` (string|null)
- `steps` (array)
- `tags` (array)
- `title` (string)
- `updatedAt` (string)
- `webUrl` (string)

### `save_my_look` (~551 tokens)

Save to My Looks

Save a new look to the user's BeautySpark My Looks. Photos can't be added from chat: if the user attached or mentions a photo, tell them before saving that only the text is saved and that they can add the photo with Add photo on the look card, when it shows one, or in BeautySpark from the returned link. Repeating the same save within 10 minutes returns the first look instead of a copy.

Input parameters:

- `categorySlugs` (array): Categories: preset slugs (everyday, work, date_night, evening, party, festival, wedding_guest, bridal, prom, holiday, vacation, photoshoot, editorial, no_makeup, smokey, glam, natural_glow, bold_colo…
- `difficultyLevel` (string): How hard the look is to do.
- `estimatedTotalTime` (integer): Time for the whole look, in minutes.
- `inspirationUrl` (string): An http(s) link to what inspired the look, kept as text. Never fetched.
- `intensityLevel` (string): How strong the whole look is.
- `isFavourite` (boolean): Mark the look as a favourite.
- `notes` (string): Free-text notes about the look.
- `occasions` (array): Free-text occasions, e.g. ["office", "wedding guest"].
- `personName` (string): Name of one of the user's persons in BeautySpark (exact name, any case). Leave it out unless the user names someone.
- `products` (array): Products used. Each: name (required); productType, e.g. "eyeshadow palette"; area; userPaletteId = the paletteId from list_my_palettes when it is a palette the user owns; shades = the pans used, each…
- `steps` (array): Ordered steps, first to last. Each step: instruction (required); title; area; positions = where on the eye, only when area is "eyes"; technique (free text); toolsNeeded = tool slugs from the allowed…
- `tagNames` (array): Tag names. A name the user already has reuses that tag; a new name creates one.
- `title` (string, required): Short name for the look, e.g. "Soft bronze smoky eye".

Output parameters:

- `created` (boolean)
- `look` (object)
- `summary` (string)

### `update_my_look` (~678 tokens)

Update my look

Change a saved look. Only the fields you send change; list fields (steps, products, tags, categories, occasions) replace the stored list whole. Look photos are kept. Steps can't be changed while any step has a photoUrl; the user edits those steps in the BeautySpark app. When the look's steps link products or shades, send productRefs and shadeRefs on every step you send ([] for a step with none): a step sent without them is refused unless products is sent too. To unlink every step, send products too.

Input parameters:

- `categorySlugs` (array): Categories: preset slugs (everyday, work, date_night, evening, party, festival, wedding_guest, bridal, prom, holiday, vacation, photoshoot, editorial, no_makeup, smokey, glam, natural_glow, bold_colo…
- `difficultyLevel` (string): How hard the look is to do.
- `estimatedTotalTime` (integer): Time for the whole look, in minutes.
- `id` (string, required): A saved look id from search_my_looks.
- `inspirationUrl`: An http(s) link to what inspired the look, kept as text. Never fetched. Send null to clear it.
- `intensityLevel` (string): How strong the whole look is.
- `isFavourite` (boolean): Mark the look as a favourite.
- `notes`: Free-text notes about the look. Send null to clear it.
- `occasions` (array): Free-text occasions, e.g. ["office", "wedding guest"]. Replaces the whole stored list, so send the complete list.
- `personName`: Name of one of the user's persons in BeautySpark (exact name, any case). Leave it out unless the user names someone. Send null to clear it.
- `products` (array): Products used. Each: name (required); productType, e.g. "eyeshadow palette"; area; userPaletteId = the paletteId from list_my_palettes when it is a palette the user owns; shades = the pans used, each…
- `steps` (array): Ordered steps, first to last. Each step: instruction (required); title; area; positions = where on the eye, only when area is "eyes"; technique (free text); toolsNeeded = tool slugs from the allowed…
- `tagNames` (array): Tag names. A name the user already has reuses that tag; a new name creates one. Replaces the whole stored list, so send the complete list.
- `title` (string): Short name for the look, e.g. "Soft bronze smoky eye".

Output parameters:

- `created` (boolean)
- `look` (object)
- `summary` (string)

### `delete_my_look` (~43 tokens)

Delete my look

Delete a saved look from the user's BeautySpark My Looks. The look is removed.

Input parameters:

- `id` (string, required): A saved look id from search_my_looks.

### `save_generation_to_my_looks` (~62 tokens)

Save a generation to My Looks

Save a finished BeautySpark generation (a jobId from list_my_generations) as a look in My Looks. Saving the same generation again returns the look already saved.

Input parameters:

- `jobId` (string, required): A jobId from list_my_generations.

Output parameters:

- `created` (boolean)
- `look` (object)
- `summary` (string)

### `get_my_profile` (~44 tokens)

Get my beauty profile

The user's BeautySpark analysis: color season, undertone, eye shape, face shape and app language. Fields are null until the selfie analysis has been run in the app.

Output parameters:

- `eyeShape`
- `faceShape` (string|null)
- `language` (string|null)
- `note` (string)
- `season`
- `undertone` (string|null)

### `list_my_palettes` (~53 tokens)

List my palettes

The palettes the user scanned in BeautySpark, with shade names and hex colors. Use them to advise with products the user owns; pass a paletteId as a product's userPaletteId in save_my_look.

Output parameters:

- `palettes` (array)

### `list_my_generations` (~65 tokens)

List my generations

The user's finished BeautySpark look generations, newest first: title, date, thumbnail and whether it is already in My Looks. Pass a jobId to save_generation_to_my_looks to save one.

Input parameters:

- `limit` (integer): How many to return (default 20).

Output parameters:

- `generations` (array)

### `try_look_from_photo` (~245 tokens)

Try a look from a photo

Prepare a try-on: BeautySpark recreates a makeup look on the user's selfie. Give one reference: articleLookId (from find_article_looks) or savedLookId (from search_my_looks). Photos (a reference photo or a different selfie) can only be uploaded in the try-on widget, never taken from chat attachments. The saved selfie is used unless the user uploads another in the widget. Returns the credit cost and balance without charging; nothing is generated until confirm_look_generation.

Input parameters:

- `articleLookId` (string): A look id from find_article_looks, to try that article look on.
- `referencePhoto` (object): Set by the try-on widget in ChatGPT. Leave unset.
- `referenceUploadId` (string): Set by the try-on widget after an upload. Leave unset.
- `savedLookId` (string): A saved look id from search_my_looks, to try one of the user's own looks on.
- `selfiePhoto` (object): Set by the try-on widget in ChatGPT. Leave unset.
- `selfieUploadId` (string): Set by the try-on widget after an upload. Leave unset.

### `confirm_look_generation` (~118 tokens)

Confirm try-on

Start the try-on prepared by try_look_from_photo. This spends the BeautySpark credits it quoted. The try-on widget calls it when the user presses Confirm. If no widget is showing, ask the user to confirm the cost first and call it only after they agree. Calling it again for the same quote returns the same job without charging again. If it answers conflict, the same quote is still being confirmed: call it again in a few seconds.

Input parameters:

- `quoteId` (string, required): The quoteId from try_look_from_photo.

Output parameters:

- `alreadyStarted` (boolean)
- `jobId` (string)

### `get_look_status` (~110 tokens)

Get try-on status

Check a try-on started by confirm_look_generation. Waits up to waitSeconds (at most 25) for it to finish before answering. When completed, the result includes the image URL and, when the look has them, its step-by-step instructions with the colours for each step.

Input parameters:

- `jobId` (string, required): The jobId from confirm_look_generation.
- `waitSeconds` (number): Seconds to wait for completion before answering, 0 to 25. Use 25.

Output parameters:

- `afterUrl` (string)
- `jobId` (string)
- `nextPollMs` (integer)
- `progress` (number)
- `status` (string)
- `step` (string|null)
- `steps` (array)
- `stepsNote` (string)

### `create_upload_url` (~41 tokens)

Create photo upload URL

Used by the try-on widget to upload a photo the user picked. Returns an upload URL valid for 15 minutes.

Input parameters:

- `contentType` (string, required)

Output parameters:

- `headers` (object)
- `uploadId` (string)
- `uploadUrl` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp#diagnostics

## Score history

- 2026-11-21: 66
- 2026-11-20: 65
- 2026-11-19: 65
- 2026-11-18: 64
- 2026-11-17: 64
- 2026-11-16: 64
- 2026-11-15: 63
- 2026-11-14: 63
- 2026-11-13: 62
- 2026-11-12: 62
- 2026-11-11: 61
- 2026-11-10: 61
- 2026-11-09: 60
- 2026-11-08: 60
- 2026-11-07: 59
- 2026-11-06: 59
- 2026-11-05: 58

## Common questions

### What is the BeautySpark MCP server?

BeautySpark is an MCP server listed in the public MCP registry as io.github.ThomasFindlay/beautyspark-mcp. BeautySpark MCP server for ChatGPT, Claude and other MCP clients. This page covers its hosted endpoint (https://mcp.beautyspark.me/mcp).

### Is the BeautySpark MCP server safe to use?

BeautySpark scores 66 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the BeautySpark MCP server expose?

BeautySpark exposes 22 tools: search_guides, get_guide, search, fetch, get_color_season, and 17 more. Their descriptions and schemas cost roughly 3,471 tokens of context every time the server is loaded.

### Does the BeautySpark MCP server require authentication?

No. We connected to BeautySpark without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the BeautySpark MCP server still maintained?

BeautySpark is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the BeautySpark MCP server?

VerifyMCP ran 14 checks on BeautySpark, covering TLS certificate, Authorization, HTTPS, HSTS header and 10 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://mcp.beautyspark.me/mcp
- Repository: https://github.com/ThomasFindlay/beautyspark
- Changelog RSS feed: https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp.json
- HTML version of this page: https://verifymcp.io/servers/thomasfindlay-beautyspark-mcp/mcp
