Skip to main content
Glama

John's Essentials

by com.johnsessentials
OverviewDiscussionsAdmin

photo_face_blur

Face Blur — Automatically detect and blur faces in a photo for privacy protection. [category: photo]

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
fileYesJPG, PNG, WebP, BMP (max 25MB)
blur_modeNogaussian softens, pixelate mosaics, solid draws an opaque black box — solid is the only mode that survives deblurring attacks.gaussian
block_sizeNoSize of each mosaic square. 0 uses the amount chosen above; 1 is ignored, use 2 or more.
blur_radiusNoGaussian radius override. 0 = auto from blur_strength.
manual_facesNoAdvanced: extra rectangles to blur even if no face was found there, as [{"x":10,"y":20,"w":80,"h":80}] in pixels of the picture as it is shown (turned upright by its EXIF tag). Decimals are rounded.
blur_strengthNoPreset intensity 1-4. blur_radius/block_size overrides beat it when set; irrelevant for solid mode.
output_formatNoOptional output format; defaults to the input format.
selected_facesNoAdvanced: which detected faces to blur, as a list of numbers starting at 0, e.g. [0,2]. Leave it blank to blur every face found; an empty list [] blurs none of them, only the areas in manual_faces. Not offered in the workflow builder: each file there has its own faces.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed10 schema fields changed
    • addedInput schema / properties / block_size / x-ui / draw
      Added value: +{
+  "from": 2,
+  "of": "areas",
+  "role": "block_size"
+}
    • addedInput schema / properties / blur_mode / x-ui
      Added value: +{
+  "draw": {
+    "fallback": "gaussian",
+    "looks": {
+      "gaussian": "blur",
+      "pixelate": "pixelate",
+      "solid": "fill"
+    },
+    "of": "areas",
+    "role": "look"
+  }
+}
    • addedInput schema / properties / blur_radius / x-ui / draw
      Added value: +{
+  "from": 1,
+  "of": "areas",
+  "role": "blur_radius"
+}
    • addedInput schema / properties / blur_strength / x-ui
      Added value: +{
+  "draw": {
+    "block": [
+      4,
+      6,
+      10,
+      15
+    ],
+    "blur": [
+      8,
+      16,
+      28,
+      42
+    ],
+    "of": "areas",
+    "role": "strength"
+  }
+}
    • changedInput schema / properties / manual_faces / description
      Previous value: -"Advanced: extra rectangles to blur even if no face was found there, as [{\"x\":10,\"y\":20,\"w\":80,\"h\":80}] in pixels."New value: +"Advanced: extra rectangles to blur even if no face was found there, as [{\"x\":10,\"y\":20,\"w\":80,\"h\":80}] in pixels of the picture as it is shown (turned upright by its EXIF tag). Decimals are rounded."
    • addedInput schema / properties / manual_faces / x-ui / draw
      Added value: +{
+  "of": "areas",
+  "role": "list"
+}
    • changedInput schema / properties / selected_faces / description
      Previous value: -"Advanced: which detected faces to blur, as a list of numbers starting at 0, e.g. [0,2]. Leave empty to blur every face found."New value: +"Advanced: which detected faces to blur, as a list of numbers starting at 0, e.g. [0,2]. Leave it blank to blur every face found; an empty list [] blurs none of them, only the areas in manual_faces. Not offered in the workflow builder: each file there has its own faces."
    • addedInput schema / properties / selected_faces / x-show-when
      Added value: +{
+  "blur_mode": [
+    "__never"
+  ]
+}
    • addedInput schema / properties / selected_faces / x-ui
      Added value: +{
+  "builder_omits": true
+}
    • addedInput schema / x-draw
      Added value: +{
+  "areas": {
+    "note": "Every face it finds is hidden the same way. Mark anything else to hide, like a name badge or a license plate.",
+    "note_no_picture": "Every face it finds is hidden the same way. Other areas can be marked only when this step works on an attached picture.",
+    "noun": "area to hide",
+    "shape": "boxes",
+    "title": "Areas to hide"
+  }
+}
  2. Changed1 schema field changed
    • addedInput schema / properties / manual_faces / x-ui
      Added value: +{
+  "unset_label": "Only the faces it finds"
+}
  3. Changed1 schema field changed
    • addedInput schema / properties / output_format / x-ui
      Added value: +{
+  "unset_label": "Keeps input format"
+}
  4. Changed2 schema fields changed
    • addedInput schema / properties / block_size / x-ui
      Added value: +{
+  "unit": "px"
+}
    • addedInput schema / properties / blur_radius / x-ui
      Added value: +{
+  "unit": "px"
+}
  5. Changed7 schema fields changed
    • changedInput schema / properties / block_size / description
      Previous value: -"Pixelate block size override. 0 = auto from blur_strength."New value: +"Size of each mosaic square. 0 uses the amount chosen above; 1 is ignored, use 2 or more."
    • addedInput schema / properties / block_size / x-show-when
      Added value: +{
+  "blur_mode": [
+    "pixelate"
+  ]
+}
    • addedInput schema / properties / blur_radius / x-show-when
      Added value: +{
+  "blur_mode": [
+    "gaussian"
+  ]
+}
    • addedInput schema / properties / blur_strength / x-show-when
      Added value: +{
+  "blur_mode": [
+    "gaussian",
+    "pixelate"
+  ]
+}
    • changedInput schema / properties / manual_faces / description
      Previous value: -"Optional JSON array of manual face rectangles to blur."New value: +"Advanced: extra rectangles to blur even if no face was found there, as [{\"x\":10,\"y\":20,\"w\":80,\"h\":80}] in pixels."
    • addedInput schema / properties / output_format / enum
      Added value: +[
+  "jpg",
+  "png",
+  "webp",
+  "bmp"
+]
    • changedInput schema / properties / selected_faces / description
      Previous value: -"Optional JSON array of detected-face indexes to blur (default: all)."New value: +"Advanced: which detected faces to blur, as a list of numbers starting at 0, e.g. [0,2]. Leave empty to blur every face found."
  6. First observed

TDQS

B3.2/5.0
Behavior3/5

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

Annotations already declare this as a non-destructive, non-read-only, closed-world operation, so the safety profile is covered structurally. The description adds that every detected face is blurred identically, which is useful behavioral context. It says nothing about permission needs, batching, or whether the source file is modified in place versus returning a new file.

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?

One sentence, front-loaded with the verb and the privacy rationale, no filler. The bracketed category tag is a minor non-informative addition but costs nothing.

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

Completeness3/5

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

For an 8-parameter image mutation tool with no output schema, the description is adequate but thin. It omits what the tool returns, whether the original is preserved, and how it relates to the adjacent detection tool, leaving the schema to carry nearly the entire burden.

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 coverage is 100% and the schema descriptions are unusually rich, including mode tradeoffs ('solid is the only mode that survives deblurring attacks'), override precedence, and the selected_faces empty-list semantics. The description adds nothing parameter-specific, so the baseline 3 applies when the schema already does this much work.

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?

States a specific verb (blur) and resource (faces in a photo) with the privacy-protection intent made explicit. Distinguishes itself from the closest sibling, photo_face_detect, which presumably detects rather than blurs, though it never names that sibling outright.

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 vs when-not guidance. In a family that contains photo_face_detect and photo_bg_remover, an agent has to infer that this is the privacy-redaction pass rather than the detection pass. No prerequisites or alternatives are stated.

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

All John's Essentials toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources