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 magic-spells/figma-mcp-bridge --skill figma-bridgegit clone --depth 1 https://github.com/magic-spells/figma-mcp-bridgeWrote 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/magic-spells/figma-mcp-bridge/figma-bridge)<a href="https://agentmods.dev/skills/magic-spells/figma-mcp-bridge/figma-bridge"><img src="https://agentmods.dev/badge/skills/magic-spells/figma-mcp-bridge/figma-bridge/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/magic-spells/figma-mcp-bridge/figma-bridge"><img src="https://agentmods.dev/badge/skills/magic-spells/figma-mcp-bridge/figma-bridge.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00000 | $0.01807 |
| Opus 5.5 | $0.00000 | $0.00723 |
| Sonnet 5.5 | $0.00000 | $0.00361 |
| Haiku 4.5 | $0.00000 | $0.00181 |
Grade A, and why
figma-bridge 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 15d 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.
How it starts
The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Figma MCP Bridge — Operation Skill
How to work through the figma-mcp-bridge effectively and honestly. Read this before any write-heavy design session. Applies to bridge v0.4.0+.
Verification discipline — the core of this skill
1. Verify bindings by readback, not by assumption.
figma_get_nodes (full depth) returns boundVariables (node-level variable
bindings), explicitVariableModes (pinned modes), layoutWrap,
counterAxisSpacing, and clipsContent. After binding, pinning, or layout
work, read the node back and confirm the field says what you intended. Mutating
tools echo readbacks in their responses (verified: true, resulting
explicitVariableModes, per-side stroke weights, etc.) — check them.
2. Export the render and look at it before reporting done.
figma_export_node writes the image to disk and returns the path — Read the
file and actually look at it. Property readback cannot see composition: real
sessions passed every numeric check while renders showed a portrait video well
where a 16:9 one belonged, a heading breaking mid-word, and a button hugging at
140px instead of spanning its form. Looking catches what measuring cannot.
3. Report honestly.
- If a tool errors, report the failure — never record the work as done.
- If a readback disagrees with a success response, report that too.
- Never paper over a missing capability by hardcoding a value.
- The bridge errors instead of silently no-oping (see error codes below); treat those errors as information about Figma's real constraints, not as obstacles to retry around.
Errors that mean "Figma forbids this" (don't retry — change approach)
INSTANCE_SUBLAYER_RESTRICTED— size binds, resizes, and reorders inside an instance are not allowed. Make the change on the component master; it flows to every instance.MODE_NOT_FOUND/COLLECTION_NOT_FOUND— the modeId/collectionId is wrong; the error lists valid modes.WRONG_EDITOR/FIGMA_DESIGN_ONLY— the tool is gated to FigJam or Figma Design; the error names the current editor type (there are five: figma, figjam, dev, slides, buzz).BIND_NOT_APPLIED/STYLE_NOT_APPLIED/RESIZE_NO_OP/REORDER_FAILED/LIMIT_NOT_APPLIED— the write did not land and the bridge is telling you instead of pretending. Report it.
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.
- 15d ago Changed · +16 lines 30d7976faa82
- 22d ago Changed · +12 lines 152c8bcf389a
- 2mo ago First seen · 94 lines · 0 tokens per session scan A afca6d577d60
figma-bridge is a skill published in the GitHub repository magic-spells/figma-mcp-bridge (84 stars, last pushed 16d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,807 tokens. 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 2026-08-30.
figma-bridge compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| figma-bridgemagic-spells/figma-mcp-bridge | Skill | 84 | 16d ago | MIT | A | 0 | Claude Code, Codex |
| webgl-holographic-foilnexu-io/open-design | Skill | 106k | today | Apache-2.0 | A | 41 | Claude Code, Codex |
| html-ppt-hermes-cyber-terminalnexu-io/open-design | Skill | 106k | today | Apache-2.0 | A | 53 | Claude Code, Codex |
| html-ppt-taste-brutalistnexu-io/open-design | Skill | 106k | today | Apache-2.0 | A | 78 | Claude Code, Codex |
| visual-ralphYeachan-Heo/oh-my-codex | Skill | 34k | 3d ago | MIT | A | 52 | Codex |
| glance-canonical-layoutsandroid/skills | Skill | 10k | yesterday | Apache-2.0 | A | 56 | Claude Code |
| accessibilitymicrosoft/hve-core | Skill | 1.6k | today | MIT | A | 47 | Claude Code |