iLook Face Analysis

Detect eye shape

detect_eye_shape

Read-only

Returns opening shape, corner orientation, and visible eye framing.

Input Schema

TableJSON Schema

Name	Required	Description
`mime_type`	Yes	Image media type. Supported formats are JPEG, PNG, and WebP.
`image_base64`	Yes	Base64-encoded JPEG, PNG, or WebP bytes, without a data URL prefix.
`adult_and_authorized`	Yes	Confirms the pictured person is an adult and the submission is lawful.

Output Schema

TableJSON Schema

Name	Required	Description
`tool`	Yes	Focused measurement key.
`error`	No	Present when the focused analysis fails.
`model`	Yes	Model version used for visible analysis.
`phi_score`	No	Optional Golden Ratio context when relevant.
`disclaimer`	Yes	Required responsible-use disclaimer.
`face_shape`	No	Optional face-shape context when relevant.
`analysis_id`	Yes	Unique analysis identifier.
`measurement`	Yes	Score, label, analysis, and tool-specific detail rows.
`rubric_version`	Yes	Versioned iLook scoring rubric.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

AddedAug 21, 2026
RemovedAug 20, 2026

Changed2 schema fields changedAug 19, 2026

addedInput schema / properties / mime_type / description

Added value: +"Image media type. Supported formats are JPEG, PNG, and WebP."

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "additionalProperties": true,
+  "properties": {
+    "analysis_id": {
+      "description": "Unique analysis identifier.",
+      "type": "string"
+    },
+    "disclaimer": {
+      "description": "Required responsible-use disclaimer.",
+      "type": "string"
+    },
+    "error": {
+      "description": "Present when the focused analysis fails.",
+      "properties": {
+        "code": {
+          "description": "Stable iLook error code.",
+          "type": "string"
+        },
+        "message": {
+          "description": "Human-readable failure explanation.",
+          "type": "string"
+        }
+      },
+      "type": "object"
+    },
+    "face_shape": {
+      "description": "Optional face-shape context when relevant.",
+      "type": "object"
+    },
+    "measurement": {
+      "description": "Score, label, analysis, and tool-specific detail rows.",
+      "type": "object"
+    },
+    "model": {
+      "description": "Model version used for visible analysis.",
+      "type": "string"
+    },
+    "phi_score": {
+      "description": "Optional Golden Ratio context when relevant.",
+      "type": "object"
+    },
+    "rubric_version": {
+      "description": "Versioned iLook scoring rubric.",
+      "type": "string"
+    },
+    "tool": {
+      "description": "Focused measurement key.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "analysis_id",
+    "rubric_version",
+    "model",
+    "tool",
+    "measurement",
+    "disclaimer"
+  ],
+  "type": "object"
+}

AddedAug 18, 2026

TDQS

B3.4/5.0

Behavior3/5

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

The readOnlyHint=true and destructiveHint=false annotations already disclose the safety profile, and the description adds specificity about extracted features (opening shape, corner orientation, framing). However, it doesn't address edge cases like closed eyes, poor lighting, multiple faces, or non-face images, which could cause unexpected output for an image-analysis tool. No contradiction with annotations found.

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

Conciseness5/5

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

A single 10-word sentence that front-loads the action verb 'Returns' and lists exactly three concrete output attributes. Zero filler, no repetition of schema or annotations — every word earns its place.

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 read-only, non-destructive image analysis tool with an output schema present and full parameter coverage, the description is reasonably complete. It states the return contents, and the output schema documents return structure. The only weakness is that terms like 'visible eye framing' are somewhat abstract without examples, and there's no guidance on input constraints (e.g., image resolution, face position) that might affect detection quality.

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% — all three parameters (image_base64, mime_type, adult_and_authorized) are fully documented with formats, enums, and purpose. The description adds zero parameter-level details, so the baseline of 3 applies; the schema carries the entire burden and does it well.

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

Purpose4/5

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

The description uses a clear verb+resource pattern ('Returns opening shape, corner orientation, and visible eye framing') and specifies three concrete output features. However, it doesn't explicitly distinguish itself from the closely-related sibling analyze_canthal_tilt, which overlaps on 'corner orientation' (canthal tilt is exactly eye corner angle), so a cautious agent might struggle to pick between them.

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

Usage Guidelines2/5

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

No when-to-use guidance, no exclusions, and no alternative tool names mentioned. Given that analyze_canthal_tilt appears in the sibling list and conceptually overlaps with 'corner orientation,' the description offers zero help in deciding which tool to invoke for eye-corner measurements.

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

All iLook Face Analysis tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.8/5.0

Disambiguation4/5

Most tools have distinct purposes (e.g., age, skin, symmetry), but 'score_face' overlaps with several others (face shape, golden ratio, attractiveness) and could be confused with the individual analyzers. Descriptions are clear enough to differentiate, but the aggregation tool introduces some ambiguity.

Naming Consistency5/5

All tools follow a consistent verb_noun pattern with lowercase and underscores (analyze_*, detect_*, calculate_*, etc.). The verb varies but the format is uniform and readable.

Tool Count5/5

14 tools is a well-scoped number for a face analysis server—not too few to be trivial, not too many to be overwhelming. Each tool covers a specific metric or feature.

Completeness5/5

The toolset covers a broad range of facial analysis dimensions (age, symmetry, ratios, skin, eye shape, hairstyle, jawline, attractiveness, color palette, golden ratio, PSL) and appears to provide a full suite for the intended domain.

Resources

Need Help?