Skip to main content
Glama
Figma Local Bridge

Figma Local Bridge

by Miart

figma_review

Save a visual, audit, and structure baseline for a Figma subtree, then compare the live node after edits to detect pixel and finding changes with a clear verdict.

Instructions

A reproducible iteration harness for one subtree: save a visual + audit + structure baseline, edit, then compare the live node and see exactly what changed. action "save" writes a local PNG plus the audit findings and the semantic fingerprint. "compare" re-renders, re-audits and re-inspects, diffs the pixels, and reports new/resolved/changed findings, severity deltas, size/fingerprint changes and a verdict: unchanged, changed, improved or regressed. A new error/warning or a size change is a regression; resolved problems are an improvement; a visual or structural difference without either is a neutral change. The document is never written: only local review files. Use figma_compare when the reference is a mockup rather than an earlier iteration.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>"
    },
    "name": "figma_review"
  }
}

Input Schema

NameRequiredDescriptionDefault
idNocompare/drop: review id; "latest" (the default) is the newest for this file
dirNoWhere reviews live; defaults to ~/.local/share/opencode/figma-bridge-reviews
nameNosave: a label for the id (defaults to the node name)
rulesNosave: audit rules; compare always uses the baseline's rules so the delta means the same thing
scaleNosave: export scale (default 1); compare: defaults to the baseline's scale
actionYessave a baseline | compare the live node against it | list | drop local files
inlineNocompare: return the heatmap as an image
nodeIdNosave: node to capture (defaults to the selection); compare: defaults to the baseline's node
fileKeyNoWhich connected Figma file to use: its fileKey (figma_whoami) or, when omitted, the only connected file. Required once more than one file is connected; figma_ping lists them.
suggestNosave: attach exact local style/variable candidates to findings
maxNodesNosave: audit node cap
toleranceNocompare: per-channel difference that still counts as equal
maxFindingsNosave: findings cap
includeHiddenNosave: audit hidden layers too

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.36.0

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the burden and does well: it discloses the critical safety fact that the document is never written (only local review files), spells out the verdict rubric (new error/warning or size change = regression; resolved = improvement; visual-only diff = neutral change), and notes compare inherits the baseline's rules and scale. The gap is the drop action, which deletes local review files — a destructive side effect never mentioned in the description and unmitigated by any annotation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The text is front-loaded with the harness concept, then the two actions, then the verdict taxonomy, then the sibling routing — a sensible order with no filler sentences. It is dense and longer than average, but almost every clause carries behavioral information an agent needs; only the verdict rules edge toward over-explanation.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 14-parameter tool with no annotations and no output schema, the description covers the interaction model, persistence model, and result interpretation well enough to call it correctly. It is incomplete on the secondary actions (list and drop appear only as schema enum labels) and on how the returned report is shaped beyond the verdict categories.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so parameter meaning is already fully documented in the schema, and the baseline is 3. The description reinforces a couple of cross-parameter semantics (compare inherits the baseline's rules and scale so deltas are comparable), but it adds little about individual parameters such as id, name, tolerance, or inline beyond what the schema states.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names concrete verbs and resources — "save a visual + audit + structure baseline", "compare the live node" — and frames the tool as an iteration harness on one subtree. It explicitly differentiates itself from the sibling figma_compare (mockup reference vs earlier iteration), so an agent can pick between them without opening schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives the intended workflow end to end (save baseline, edit, compare) and states the exclusion explicitly: use figma_compare when the reference is a mockup rather than an earlier iteration. That is a clear when-to-use plus a named alternative.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.