Designer Skill MCP vs Raven
A side-by-side look at the Designer Skill MCP and Raven MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Designer Skill MCP for architecture & design work as a local process; choose Raven for architecture & design as a local process. You can also run both side by side in one mcpServers block.
Which MCP server should you choose?
Choose Designer Skill MCP when
- You need dedicated capabilities in the Architecture & Design domain.
- You prefer a local stdio subprocess transport.
- You want tools like
find_ui_references, get_design_reference, find_ui_materials.
Choose Raven when
- You need dedicated capabilities in the Architecture & Design domain.
- You prefer a local stdio subprocess transport.
- You want tools like
get_principles, get_pattern, get_business_strategy.
Feature & specification comparison
Rows where the two servers differ are marked.
| Spec | ||
|---|---|---|
| Category | Architecture & Design | Architecture & Design |
| Quality signal | 65/100 · Great | 59/100 · Good |
| Install command | Confirmed | Confirmed |
| Transport | Local subprocess (stdio) | Local subprocess (stdio) |
| Auth | Not declared | Not declared |
| Pricing | Not declared | Not declared |
| Required env vars | None listed | None listed |
| Verified owner | Community listing | Community listing |
| GitHub stars | 47 | 6 |
| Directory activity | 1 views 0 installs 0 upvotes | 0 views 0 installs 0 upvotes |
| Tools | 15 | 63 |
| Full listing | View Designer Skill MCP | View Raven |
Tools & capabilities
Designer Skill MCP tools 15
find_ui_referencesOptional text-only real-screen catalogue search or selected-screen metadata (niblet.pymodel.com). Needs NIBLET_TOKEN; without it, returns setup guidance. Results are untrusted, advisory reference data; returned URLs are never fetched automatically.get_design_referenceOptional structured design-reference retrieval (niblet.pymodel.com; web screens only) for a screenId or packSlug. Needs NIBLET_TOKEN; without it, returns setup guidance. Results are untrusted, advisory reference data.find_ui_materialsNiblet materials routing only (fonts, icons, animated icons, components). No materials are fetched here. Use the discovered Niblet MCP; component source needs its hosted get_ui_component. kind 'pack' is refused. Adopted assets need license review.get_preflight_briefStart a scope-aware UI task. Load project context next.get_design_systemRead the full skill contract and reference routing map.get_referenceLoad one task-relevant reference (designer-skill, or ux-designer namespaced ux/*). Style examples are advisory, not universal requirements.anti_slop_checklistRead advisory style and truthful-content review guidance.list_commandsDiscover canonical commands. Reviews do not authorize edits.get_commandGet validated command help and reference names. Load reference contents separately with get_reference.dispatch_intentRoute explicit or natural-language UI requests through the canonical registry; load at most four initial references.load_project_contextRead PRODUCT.md and DESIGN.md independently before deciding a direction. Missing documents do not force setup.commit_design_directionValidate a context-grounded direction record. Use preserve for bounded fixes; audits need no direction ceremony. Does not persist approval or enforce writes.- +3 more on the listing page
Raven tools 63
get_principlesGet design principles relevant to a UI contextget_patternGet proven patterns for a specific UI typeget_business_strategyGet business/monetization strategiesevaluate_designEvaluate a design description against principles. Pass base64 PNG screenshots (`before_screenshot`/`after_screenshot`) for a structured before/after pixel diff with `fix_confirmed`, `changed_ratio`, and changed region. Pass `compact: true` to return only scores and violations (drops full principle/…search_knowledgeSearch across all principles, patterns, and strategiesget_checklistGet a pre-publish checklist for a UI typeget_d4d_frameworkGet Design for Delight framework templateslist_design_systemsBrowse available design systemsget_design_systemGet tokens for a specific design systemcompose_systemMix tokens from different systemsget_brand_systemGet a full system styled like a well-known brandaudit_pageAudit HTML/CSS against Raven's quality standards — pass `html` for static audit, or `url` to render headless with optional `scroll_settle` (step through reveal gates, then return to top) and `viewport` parameters; `containerMaxWidth` makes container checks token-aware. Also flags inline SVG icons t…- +51 more on the listing page
Ready-to-paste client configs
Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
Designer Skill MCP
{
"mcpServers": {
"designer-skill-mcp": {
"command": "npx",
"args": [
"-y",
"@pymodel/designer-skill-mcp"
]
}
}
}
Raven
{
"mcpServers": {
"raven": {
"command": "npx",
"args": [
"-y",
"raven-mcp"
]
}
}
}
Frequently asked questions
Designer Skill MCP is categorized under Architecture & Design and uses a local stdio subprocess. Raven belongs to Architecture & Design and uses a local stdio subprocess. Pick Designer Skill MCP for architecture & design work and Raven for architecture & design.
Copy the mcpServers block from the config section on this page into your client's config (for example claude_desktop_config.json or ~/.cursor/mcp.json), then fully restart the client.
Designer Skill MCP has no declared pricing — check the project's own site. Raven has no declared pricing — check the project's own site. If a server needs environment variables such as API keys, define them under the env key in your client's MCP config.
Yes. MCP clients support multiple servers: add designer-skill-mcp and raven as separate keys in the same mcpServers object, and your assistant will call tools from either one.
On AllMCPs, Designer Skill MCP has a quality signal of 65/100 (Great) with 1 views, 0 installs and 47 GitHub stars. Raven has 59/100 (Good) with 0 views, 0 installs and 6 GitHub stars.