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 agentmods add skills/adilkalam/orca/adobe-executionnpx skills add adilkalam/orca --skill adobe-executiongit clone --depth 1 https://github.com/adilkalam/orcaWhat 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 | $0.00042 | $0.03246 |
| Opus 5 | $0.00021 | $0.01623 |
| Sonnet 5 | $0.00008 | $0.00649 |
| Haiku 4.5 | $0.00004 | $0.00325 |
Grade A, and why
adobe-execution 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 2d 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 — 369 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adobe Execution Skill
RULE: Never place without measuring. Never declare done without looking.
Core Principle
Every element you create has a position. Every position is verifiable. You MUST verify positions after placement, not assume they are correct. The loop is: measure canvas -> plan positions -> place element -> verify position -> correct if wrong -> repeat -> visually review -> fix issues -> done.
Mental Model
Think of a canvas as a spatial budget. Before placing anything, you need to know:
- Canvas dimensions -- How much space exists (width, height)
- Existing elements -- What is already placed and where
- Margins -- Safe zones from edges (proportional, not fixed)
- Relationships -- How elements relate spatially (centered, stacked, grouped)
Every placement decision is a proportion of the canvas, not a magic number.
Rule 1: Measure Before You Place
Read the canvas dimensions and existing layer positions BEFORE creating any element.
Photoshop
WRONG:
create_single_line_text_layer({ text: "Title", position: { x: 100, y: 50 }, fontSize: 48 })
create_single_line_text_layer({ text: "Subtitle", position: { x: 100, y: 120 }, fontSize: 24 })
// Where did 100, 50 come from? What if the canvas is 400px wide? 4000px wide?
RIGHT:
// Step 1: Read canvas
get_document_info()
// Returns: { width: 1200, height: 800, ... }
// Step 2: Calculate proportional positions
// margin = width * 0.05 = 60px
// title_x = margin = 60
// title_y = height * 0.08 = 64
// subtitle_y = title_y + fontSize + (height * 0.02) = 64 + 48 + 16 = 128
// Step 3: Place with calculated values
create_single_line_text_layer({ text: "Title", position: { x: 60, y: 64 }, fontSize: 48 })
create_single_line_text_layer({ text: "Subtitle", position: { x: 60, y: 128 }, fontSize: 24 })
Illustrator
WRONG:
execute_extend_script({
script: `
var doc = app.activeDocument;
var text = doc.textFrames.add();
text.contents = "Title";
text.position = [100, 700];
`
})
// 100, 700 means nothing without knowing the artboard size
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.
- 2d ago First seen · 369 lines · 42 tokens per session scan A 1e9c2fa92eca
adobe-execution is a skill published in the GitHub repository adilkalam/orca (2 stars, last pushed 1mo ago), licensed MIT. It adds 42 tokens to every session and 3,246 once invoked, about $0.0002 per session on Opus 5. 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 2026-08-31.
Other skills, from other repositories
design-guide
Paperclip UI design system guide for building consistent, reusable frontend components. Use when creating new UI components, modifying existing ones, adding pages or features to the frontend, styling UI elements, or when you need to understand the design language and conventions. Covers: component creation, design…
gstack-openclaw-office-hours
Use when asked to brainstorm, evaluate whether an idea is worth building, run office hours, or think through a new product idea or design direction before any code is written.
accessibility
Consolidated accessibility skill entrypoint for WCAG 2.2, ARIA Authoring Practices, cognitive accessibility, Section 508, EN 301 549, design intent verification, and the Accessibility Planner workflow.
deck-course-module
暖纸背景 + Playfair, 左侧学习目标常驻, 含 MCQ 自测页.
desktop-principles
Desktop-specific UX principles - hover states, pointer precision, keyboard shortcuts, multi-window, focus management. Covers macOS, Windows, Linux, web desktop.
ios-hig-design
Design native iOS interfaces following Apple Human Interface Guidelines. Use when the user mentions "iPhone app", "iPad layout", "SwiftUI", "UIKit", "Dynamic Island", "safe areas", "HIG compliance", "SF Symbols", "haptic feedback", "iOS accessibility", "make my app feel native", or "follow Apple design guidelines".…