- Home
- Skills
- Alekspetrov
- Navigator
- Product Design
product-design_skill
- Python
142
GitHub Stars
6
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill alekspetrov/navigator --skill product-design- GETTING-STARTED.md2.3 KB
- INSTALL.md7.7 KB
- README.md8.1 KB
- requirements.txt289 B
- setup.sh2.9 KB
- SKILL.md19.8 KB
Overview
This skill automates design review, token extraction, component mapping, and implementation planning to shrink design handoff time from hours to minutes. It connects directly to Figma via the MCP protocol or accepts manual design input, then generates analysis, a phased implementation plan, and task artifacts ready for development. Auto-invokes when users mention design review, Figma mockups, or design handoff requests.
How this skill works
The skill inspects Figma metadata, variables, and component structures via a local MCP connection or manual JSON input and extracts design tokens, component definitions, and similarity scores. It runs a codebase audit to detect drift, maps Figma components to code components (with confidence levels), and produces a phased task document and design-review report. Outputs include DTCG-formatted tokens, component mappings, an audit summary, and a Navigator-style implementation plan ready for PM or developer use.
When to use it
- When you need a fast design-to-code handoff for a new feature
- When you want to extract or sync design tokens from a Figma file
- When checking for design-system drift between Figma and the codebase
- When planning phased implementation and acceptance criteria for UI work
- When you need a reproducible design review report for stakeholders
Best practices
- Run the MCP automated workflow with the Figma Desktop app and MCP enabled for fastest, most accurate results
- Keep a .agent/design-system UI kit and token inventory committed to the repo for reliable mapping and diffs
- Provide a clear feature name and Figma URL or a concise manual JSON describing tokens and components when MCP is unavailable
- Review generated task documents before auto-starting implementation and mark breaking changes clearly
- Set up visual regression testing after implementation to prevent future drift
Example use cases
- Perform a full design handoff for a dashboard redesign and get a TASK document plus a review report in minutes
- Audit an existing UI for token drift and receive a prioritized list of token fixes and reuse opportunities
- Map new Figma components to existing code components and get confidence scores for reuse vs. new creation
- Generate a phased implementation plan (tokens → atoms → molecules → organisms) ready for PM ticket creation
- Extract DTCG-format tokens from Figma variables to update theme and Tailwind configs
FAQ
Figma Desktop running with MCP enabled, Python dependencies installed via the setup script, and the project structure including .agent/design-system present.
Can I use this without Figma MCP?
Yes. Provide manual design inputs or a JSON export describing tokens and components; the skill supports a manual workflow with the same analysis and planning outputs.
How accurate are component mappings?
Mappings use code-connect data first (100% confidence when available), then fuzzy name matching; each mapping includes a confidence score and unmapped items are flagged for creation.