tattoo-brief-mcp
# tattoo-brief-mcp
Official MCP Registry: `io.github.MazzaWill/tattoo-brief-mcp` · Also on [Smithery](https://smithery.ai/server/eraplug/tattoo-brief-mcp) and [Glama](https://glama.ai/mcp/servers/Stark-Will/tattoo-brief-mcp).
Related Tatany assets: [website](https://tatany.app) · [AI tattoo generator](https://tatany.app/ai-tattoo-generator) · [YouTube](https://www.youtube.com/@tatanyapp) · [itch.io flash pack](https://lenahart.itch.io/tattoo-flash-line-art-pack) · [Hugging Face probes](https://huggingface.co/datasets/yuzhucong/tattoo-probes) · [Pinterest board](https://www.pinterest.com/lenahart_ink/tattoo-flash-line-art/)
A small [Model Context Protocol](https://modelcontextprotocol.io) server that helps you plan a tattoo before you talk to an artist. It runs locally over stdio, needs no API keys, and every tool is deterministic (same input, same output).
Maintained by Lena Hart (ZheLogic, LLC). Homepage: https://tatany.app
## Tools
| Tool | What it does |
|---|---|
| `build_tattoo_brief` | `idea`, `style`, `placement`, `size_cm` (+ optional `color`, `must_include`, `avoid`) → a Markdown brief with placement notes and questions to ask your artist. |
| `check_small_size_legibility` | `size_cm` + `details` (e.g. `fine_lines`, `small_text`, `dotwork`, `dense_shading`, `tiny_negative_space`, `faces_portrait`, `color_gradients`, `white_ink`, `many_small_elements`, `bold_outline`) → per-detail blur risk and a suggested minimum size. |
| `line_art_prompt` | `subject` (+ optional `style`, `line_weight`, `composition`, `extra`) → a clean line-art prompt and negative prompt for an AI tattoo generator. |
**Important:** the legibility check uses simple rules of thumb. It is not a guarantee about how a tattoo will heal or age, and nothing here is medical advice. Your tattoo artist has the final say.
## Install
```bash
npm install && npm run build
```
Claude Desktop / Cursor config:
```json
{ "mcpServers": { "tattoo-brief": { "command": "node", "args": ["/path/to/tattoo-brief-mcp/dist/index.js"] } } }
```
Test all three tools: `npm test`. Or inspect interactively: `npx @modelcontextprotocol/inspector node dist/index.js`.
## License
MIT
TDQS
Scored across 3 tools
line_art_prompt and build_tattoo_brief both take a tattoo idea and return text, so an agent could momentarily hesitate over which produces the AI prompt versus the human-facing brief. Their descriptions do clarify the distinction (generator prompt + negative prompt vs. structured artist brief), and check_small_size_legibility is clearly separate.
All three names use snake_case, and two follow a verb_noun/verb_phrase pattern (build_tattoo_brief, check_small_size_legibility). line_art_prompt is a noun phrase rather than a verb-led name, a minor deviation from an otherwise consistent convention.
Three focused tools fit a narrow, well-defined pipeline (idea → brief → prompt → legibility check) with no filler. It sits at the thin end of the suggested range, but each tool has a distinct role, so it is not under-scoped.
The surface covers generating a prompt, building a brief, and a size-legibility heuristic, but there is no way to refine or update an existing brief/prompt after feedback, and no style or placement reference lookup. Agents can work around this by re-calling tools from scratch, but iteration is a real gap.