Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add neozhehan/figma-edit-mcp --skill figma-editgit clone --depth 1 https://github.com/neozhehan/figma-edit-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/neozhehan/figma-edit-mcp/figma-edit)<a href="https://agentmods.dev/skills/neozhehan/figma-edit-mcp/figma-edit"><img src="https://agentmods.dev/badge/skills/neozhehan/figma-edit-mcp/figma-edit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/neozhehan/figma-edit-mcp/figma-edit"><img src="https://agentmods.dev/badge/skills/neozhehan/figma-edit-mcp/figma-edit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.00438 |
| Opus 5.5 | $0.00033 | $0.00175 |
| Sonnet 5.5 | $0.00016 | $0.00088 |
| Haiku 4.5 | $0.00008 | $0.00044 |
Grade A, and why
figma-edit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
figma-edit
figma-edit-mcp lets an agent read and edit a live Figma document through a plugin that enforces safety constraints at execution time. Two things to internalize before acting:
- Discover before acting. Every workflow that touches a node starts with a read (
page_info→node_info). IDs and names the user gives you may be stale; verify them. - The plugin can refuse you. Hard constraints (scope, name verification, per-item batch checks) are checked inside Figma and return structured error codes. You cannot bypass them — plan calls to comply up front.
- Explicit variable scopes. When creating a new variable, you must explicitly provide its
scopes.
References
Load the one you need:
- references/constraints.md — the hard rules the plugin enforces (scope locking, name verification, no implicit selection, batch validation, URL IDs) and what to do when a constraint forbids the request.
- references/error-playbook.md — every structured error code and its correct recovery.
- references/workflows.md — the discover-before-acting pattern and canonical recipes (find-and-update text, create-in-frame, edit-by-URL, bulk text replacement).
- references/tool-selection.md — choosing among overlapping tools,
node_infofield selection (fast vs slow), batch vs single, and streaming progress.
The same content is available over the MCP connection as resources under figma-edit://guide/*.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 30 lines · 82 tokens per session scan A 4d4f2b14541a
figma-edit is a skill published in the GitHub repository neozhehan/figma-edit-mcp (14 stars, last pushed 8d ago), licensed MIT. It adds 82 tokens to every session and 438 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
design
Design and critique any visible UI, before and after building. Trigger on UI work: "improve the UI", "redesign", "polish", "make it look better", "visual upgrade", "design review", "design audit", "layout fix", "UX improvement", or any task whose deliverable is a page, screen, or component that renders. Shapes a…
zenith
Author and modify editable .zen designs with the Zenith CLI. Use for posters, slides, diagrams, charts, brand kits, size variants, and CSV mail-merge. Export PNG, self-contained SVG with outlined text, or PDF with explicit raster fallback reports. Use other tools for photo generation, raster editing, and unrelated…
html-design-revision
Revise an existing HTML/CSS design file with small, safe edits. Use when asked to change a color, text, size, position or add an element to a design that already exists. Use other tools for building a design from scratch.
html-author
Build a finished one-page or multi-page design (flyer, letterhead, handout, deck, poster, menu, card, report, invoice) as a single HTML/CSS file at a fixed pixel size. Use when asked to design something from a brief, starting from nothing. Use other tools to edit an existing file.
solidworks-mcp
Build, edit, inspect and export native SolidWorks models through solidworks-mcp. Use for CAD modeling, raster-to-sketch work and SolidWorks COM/MCP diagnostics. Read before calling a SolidWorks MCP tool.
site-header
Design and build the site-wide header (logo, menu, language switcher, cart, call to action) yourself, then have the site save it in the right places and check it. Activate whenever the user asks to create, fix, redesign, modernise, tidy or "make professional" a header or top navigation, or complains that the menu…