- Home
- Skills
- Branding & Design
- Figma Canon
Works with the AI Agents you already use
Figma Canon
Read-only Figma canon for AI agents: 15 on-demand references and 11 hard rules for Plugin API writes, auto layout, tokens and handoff.
Free
About this skill
Read-only knowledge base of Figma canon for AI agents. A routing table tells the agent which references to load for each task. It also carries 11 always-on hard rules learned in production, such as: the canvas mutates under you, null-guard before operating, token instead of hex. It informs writes and never performs one. Knowledge only: no scripts, no network, no credentials.
What the 15 references cover:
- Plugin API core, data and anomalies, write atomicity, auto layout
- Inspection protocol, handoff format, naming, state coverage
- Quality rubric, AI slop signatures, security, rate-limit recovery
- Code Connect setup and dated field notes (90 gotchas in the anomalies and field-notes files)
Written for figma-console-mcp (Southleft) with its Desktop Bridge plugin in Figma Desktop, or the official Figma MCP server. The other figma-maxxing skills link to its references, so install it with them.
Part of figma-maxxing, free and MIT on GitHub: https://github.com/thiagoxikota/figma-maxxing
The skill is instructions the agent reads. It does not enforce permissions or block a write, so draft-first is an instruction, not a check. Write path B (use_figma on the official Figma MCP server) is not validated by the repo: keep writes small and read back every change. figma-console-mcp was last checked at v1.40.8, and the Plugin API names the skills cite were checked against @figma/plugin-typings 1.140.0. Two field notes point to the mcp-direct scripts (fx.py, shot.py) that ship in the figma-bridge-doctor skill, not in this one.
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
Markdown only: SKILL.md plus 15 reference files, no scripts. The agent reads the references the routing table names for the task, and the project's Figma map (default docs/figma-map.md) when it exists; it writes nothing. FIGMA_ACCESS_TOKEN is named only to tell the agent never to print it, write it to a file, ask for it or commit it. Some references contain shell samples as documentation (npm install --save-dev @figma/code-connect, npx figma connect publish --token=$FIGMA_ACCESS_TOKEN, a GitHub Actions line with ${{ secrets.FIGMA_ACCESS_TOKEN }}, curl examples); the skill runs none of them. SKILL.md also names an optional precheck hook, hooks/figma-canon-precheck.py, which sits at the root of the figma-maxxing repo, is opt-in and does not ship in this skill.