---
title: "pil-agent-plugin"
description: "Quantitative image measurement plus optional constrained multi-view reconstruction and Blender geometry checks."
canonical: https://agentpluginsdirectory.com/plugins/pil-agent-plugin
last-updated: 2026-10-02
---

# pil-agent-plugin
Quantitative image measurement plus optional constrained multi-view reconstruction and Blender geometry checks.
- Slug: pil-agent-plugin
- Publisher: bsmi021
- Repository: https://github.com/bsmi021/pil-agent-plugin
- Manifest: plugin.json
- Version: 0.10.0
- License: MIT
- Category (editorial): creative-media
- Skills: 4 (bootstrap, image-analysis, image-measurement, multiview-reconstruction)
- MCP servers: 0
- Stars: 2
- Repository created: 2026-08-18
- Repository last pushed: 2026-10-01
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/pil-agent-plugin
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What pil-agent-plugin does, in the publisher's words

An agent plugin that gives coding agents quantitative image measurement, exact colour palettes, per-hue census, layout statistics, perceptual hashes, and changed-region localisation, plus an optional constrained multi-view layer for template-mesh fitting, Blender BVH clearance, and arbitrary matched renders.

It is designed to complement an agent's native multimodal vision, not replace it.

An agent with multimodal vision already reads images well: it transcribes text, understands layout semantically, identifies objects, and recognises visual style. What it cannot do is produce a number, an exact hex value, a reproducible similarity score, or coordinates of what changed between two renders.

- Native vision, unassisted, produced a full text transcription, semantic layout, object inventory, correct style identification, and a correct reading of the image's colour-coded semantics.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/bsmi021/pil-agent-plugin/HEAD/README.md

## Skills

- bootstrap: Install or diagnose PIL Agent Plugin dependencies on Windows, macOS, and Linux, and check whether bootstrap has already completed for the current plugin environment. Use for initial plugin setup, missing dependencies, or bootstrap status requests.
- image-analysis: Combines native visual inspection, quantitative Pillow image measurement, calibrated multi-view reconstruction constraints, Blender BVH geometry checks, and matched-render review. Use when an image-analysis request spans several evidence layers, such as concept-to-model comparison, seven-view chara…
- image-measurement: Measures images numerically and compares two images with Pillow, a one-call maximal profile per image (file facts, exact hex colour palettes with coverage, per-hue census, luminance/saturation/entropy statistics, tonal percentiles and clipping, per-region grid statistics, persistable perceptual-ha…
- multiview-reconstruction: Prepares calibrated multi-view concept images, fits a template mesh with OpenCV and SciPy constraints, checks Blender garment clearance with BVH, renders arbitrary locked-framing views, and performs worst-case image review. Use for reconstructing or fitting an existing 3D mesh from several image vi…

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.