- Home
- Skills
- Branding & Design
- Figma Slop Check
Works with the AI Agents you already use
Figma Slop Check
Gate for Figma frames after a write: slop and rigor checks, then PASS or a numbered punch list by node id.
Free
About this skill
Quality gate for a Figma file, run after a write and before figma-handoff-gate. No frame is called done while a check fails. It reads the file and reports PASS, or a numbered punch list by node id where each finding carries a severity and an action (swap, snap or ask). Rigor findings stay read-only until the user approves them.
What it checks:
- Slop, 8 checks (S1 to S8): overlap and spacing, AI-stuffed cards, status marker creep, names and canvas notes that read as generated, decorative excess, mechanical repetition, hype copy, and the closing reply.
- Rigor, 11 checks in 5 groups: bound or loose (color, text style, scale), components intact, auto layout sized right, words (glossary wording, layer names), and screenshot proof.
- The 5 auto-fails and the 0-10 score of the figma-canon quality rubric, 8/10 minimum.
Requires the figma-canon skill installed next to it (it links to ../figma-canon/references/). Written for figma-console-mcp with the Desktop Bridge plugin in Figma Desktop; not validated on the official Figma MCP server.
Part of figma-maxxing, free and MIT on GitHub: https://github.com/thiagoxikota/figma-maxxing
Not validated on the official Figma MCP server; the checks are written for figma-console-mcp and its Desktop Bridge plugin. The repo README reports one demo run on that server, made with the rigor lens from before 1.2.0, which rewrote it as the current 11 checks. Depends on figma-canon for the slop catalog, quality rubric, naming, state and handoff canon. Scales, glossary and naming rules come from the project's variables, styles and figma-map.md; the numbers in references/checks.md are placeholders, and a case no source covers is reported as ask. With two or more variable modes, values are read in the current mode only. Accessibility beyond the catalog's auto-fails and layout strategy are out of scope. If the Bridge is disconnected it stops and hands off to figma-bridge-doctor, a separate skill. After 3 fix and re-check cycles on the same failing check it stops and surfaces the problem.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
2 uses
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts
File Scopes
The skill folder holds Markdown only (SKILL.md, references/checks.md, references/punch-list.md) and ships no scripts. It makes no network calls and reads no environment variables itself; Figma is reached through the agent's figma-console-mcp tools. It reads the project map (default docs/figma-map.md) and the figma-canon references. It writes one run record per screen to .figma-slop-check/runs/<YYYY-MM-DD>-<screen-slug>.json and keeps accepted deviations in .figma-slop-check/decisions.json. Canvas changes happen only for fixes the user approves, each routed through figma-preflight, plus slop failures on work the agent just made, which it is told to fix before answering.