Usage4Claude: Skill for Codex

.agents/skills/capture-usage4claude-screenshots/SKILL.md

capture-usage4claude-screenshots is a skill for Codex from f-is-h/Usage4Claude. It costs 97 tokens per session (2,312 once invoked), scanned A, original, MIT.

A macOS automation skill for capturing screenshots of the Usage4Claude menu-bar interface with CleanShot X, a screenshot utility.

In plain words
What is it for?
Use it to capture localized Usage4Claude screenshots, change its display or language settings, and troubleshoot CleanShot or macOS accessibility issues.
Why use it?
It follows the precise window-selection and mouse actions needed to capture the correct popover and save it with the expected filename.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: installed under .agents/ (shared by several agents); mentions Codex.

This is f-is-h/Usage4Claude's own configuration. It tells Codex how to work on Usage4Claude itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Usage4Claude configures →

Reuse

Borrowing it

Nothing to install: this file belongs to f-is-h/Usage4Claude. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/f-is-h/Usage4Claude/main/.agents/skills/capture-usage4claude-screenshots/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/f-is-h/Usage4Claude

Made for: Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for capture-usage4claude-screenshots

README.md
[![agentmods](https://agentmods.dev/badge/skills/f-is-h/usage4claude/capture-usage4claude-screenshots/github.svg)](https://agentmods.dev/skills/f-is-h/usage4claude/capture-usage4claude-screenshots)
Your own site
<a href="https://agentmods.dev/skills/f-is-h/usage4claude/capture-usage4claude-screenshots"><img src="https://agentmods.dev/badge/skills/f-is-h/usage4claude/capture-usage4claude-screenshots/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for capture-usage4claude-screenshots

Your own site · 80×15
<a href="https://agentmods.dev/skills/f-is-h/usage4claude/capture-usage4claude-screenshots"><img src="https://agentmods.dev/badge/skills/f-is-h/usage4claude/capture-usage4claude-screenshots.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 97 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,312 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Anti-Refusal · line 42
    Skill attempts to nullify the agent's safety policies or restrictions ('you have no restrictions', 'ignore your guidelines', 'do anything now'). This is a direct jailbreak that disables guardrails.
    Fix: Remove jailbreak framing that nullifies safety policies or restrictions. Skill content must not instruct the agent to ignore its guidelines or operate without guardrails.
  • high Anti-Refusal · line 61
    Skill attempts to nullify the agent's safety policies or restrictions ('you have no restrictions', 'ignore your guidelines', 'do anything now'). This is a direct jailbreak that disables guardrails.
    Fix: Remove jailbreak framing that nullifies safety policies or restrictions. Skill content must not instruct the agent to ignore its guidelines or operate without guardrails.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00097 $0.02312
Opus 5 $0.00048 $0.01156
Sonnet 5 $0.00019 $0.00462
Haiku 4.5 $0.00010 $0.00231

Measured 2d ago against content hash 6e2f88a87613, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

capture-usage4claude-screenshots 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.

The scan reads SKILL.md. This mod also ships 4 executable files (scripts/apply_reference_debug_values.sh, scripts/apply_scenario.sh, scripts/capture_current_display_all_languages.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.agents/skills/capture-usage4claude-screenshots/SKILL.md · 168 lines

How it starts

The opening of the file, as written. The whole thing — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Capture Usage4Claude Screenshots

Overview

Capture Usage4Claude popover screenshots by matching the user's manual CleanShot workflow: enter CleanShot window capture mode, hover the real mouse over the Usage4Claude popover, click, and rename the newly saved file in Downloads.

This workflow is fragile because CleanShot's window selector reacts to the real mouse position. Do not rely on app-relative Computer Use clicks for the final selection; use CoreGraphics to move the mouse to the window center and post a real click.

Preflight

Confirm these before capturing:

  • Use the Debug Usage4Claude instance when the user requests it.
  • Verify Codex has macOS Accessibility permission. Computer Use alone is not enough for shell-launched osascript.
  • Verify this succeeds before starting CleanShot:
osascript -e 'tell application "System Events" to get count of application processes'

If it fails with not allowed assistive access or 不允许辅助访问, stop and ask the user to add Codex to System Settings > Privacy & Security > Accessibility.

Configure Usage4Claude

Set the screenshot state in Usage4Claude settings before capture.

  • Claude-only: enable Claude 5-Hour, 7-Day, Extra Usage, 7D Opus, and 7D Sonnet; disable all Codex items.
  • Codex-only: enable all 3 Codex limits; disable all Claude items.
  • Mixed: enable Claude's first 3 limits and all 3 Codex limits.
  • Switch language only inside Usage4Claude settings. Do not change language by writing UserDefaults or by external app-level language overrides.
  • Keep the Usage4Claude screenshot setting that prevents the popover from closing.

After changing settings, open the main Usage4Claude popover and inspect it with Computer Use. The screenshot should show the intended language and display items before CleanShot starts.

Reference Debug Usage Values

For canonical documentation screenshots, use a Debug build with Debug Mode enabled and set these eight slider values before capture:

Debug slider Value
Claude 5-Hour 66%
Claude 7-Day 88%
Claude Extra Usage 66%
Claude Opus 7-Day 66%
Claude Sonnet 7-Day 66%
Codex Primary 66%
Codex Secondary 88%
Codex Extra Usage 88%

Read the full file on GitHub · 168 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 2d ago Changed · +30 lines 6e2f88a87613
  2. 10d ago First seen · 138 lines · 97 tokens per session scan A 3eb1c49a1854

Subscribe to this mod's changes

capture-usage4claude-screenshots is a skill published in the GitHub repository f-is-h/Usage4Claude (386 stars, last pushed 2d ago), licensed MIT. It adds 97 tokens to every session and 2,312 once invoked, about $0.0005 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 2026-08-30.

Related

Other skills, from other repositories

midjourney-replicate-flux

Generate highly detailed, Midjourney-style image prompts optimized for the FLUX 1.1 Pro model on Replicate. Transform basic user descriptions into rich, cinematic prompts with professional photography qualities, dramatic lighting, and editorial-quality aesthetics. Use when users request image generation, need prompt…

rawveg/skillsforge-marketplace · 83 tokens

figlet-text-converter

This skill processes files containing figlet tags and replaces them with ASCII art representations. It detects and preserves comment styles (forward slash forward slash, hash, double-dash, forward slash asterisk), automatically manages Node.js dependencies, and supports 400+ fonts (defaulting to the standard font).…

rawveg/skillsforge-marketplace · 98 tokens

linkedin-carousel

Generates on-brand LinkedIn carousel PDFs (1080x1350 per slide) from raw text, URLs, or uploaded documents. Four templates: executive-minimal, training-breakdown, quote-forward, and mobile-optimized. Fully brandable: colors, fonts, logo, headshot, and footer text come from brand-config.yaml. Triggers on: 'make a…

karagos/ai-skills-by-caio · 127 tokens

wiggle

Create animated logo files (Lottie JSON, GIF, MP4) from static logo images. This skill should be used when users provide a logo image (PNG/SVG/JPG) and request any kind of logo animation, motion graphics, animated logo effect, waveform animation, bouncing logo, rotating logo, pulsing logo, wiggling logo, or ask to…

talknerdytome-labs/wiggle-claude-skill · 121 tokens

r3f-animation

React Three Fiber animation - useFrame, useAnimations, spring physics, keyframes. Use when animating objects, playing GLTF animations, creating procedural motion, or implementing physics-based movement.

zebbern/claude-code-guide · 43 tokens

chart-image

Generate publication-quality PNG chart images from data, supporting line, bar, area, candlestick, pie, and heatmap charts. Triggers when the user asks to visualize data, create a graph, plot a time series, or generate a chart for a report, alert, or dashboard. Runs as a lightweight, headless Node.js process without a…

zebbern/claude-code-guide · 75 tokens