메뉴
macOS-only Raycast script to paste clipboard images into a terminal-based AI agent as file attachments. Use when the user wants to paste/attach a screenshot or clipboard image to a terminal agent like Claude Code. Requires macOS with Raycast installed.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Create, modify, or inspect Slack apps and bots via the Manifest API. Use when adding slash commands, updating scopes, changing bot settings, or creating new Slack apps on the swabby workspace.
Store, encrypt, decrypt, and rotate API keys, tokens, and credentials locally using AES-256-GCM encryption. Use this skill whenever the user mentions API keys, secrets, tokens, credentials, passwords, service accounts, or needs to securely store/retrieve sensitive values. Also triggers when other skills (cloud-provisioning, github, gitlab) need to persist credentials after onboarding.
Switch which Anthropic auth profile (OAuth token) OpenClaw uses. Use when the user asks to switch API keys, swap auth tokens, change which account is active, check which profile is in use, or troubleshoot rate limiting by switching to the backup account.
Onboard and provision compute credentials for AWS, Azure, and GCP. Use this skill whenever the user needs to set up cloud credentials, create IAM users or service accounts, configure service principals, or provision compute access on any major cloud platform. Walks through step-by-step onboarding with least-privilege policies and stores credentials via the secret-vault skill.
Browse and search a curated index of AI agent skills from a GitHub repo. Use when the user wants to find, list, browse, or install skills, or asks what skills are available.
Scaffold project environments with directory structure, configuration files, .env templates, and dependency manifests based on project type. Use this skill whenever the user wants to bootstrap a new project, create a project skeleton, scaffold an environment, set up a new repo structure, or generate boilerplate configuration.
| name | clip-img |
| description | macOS-only Raycast script to paste clipboard images into a terminal-based AI agent as file attachments. Use when the user wants to paste/attach a screenshot or clipboard image to a terminal agent like Claude Code. Requires macOS with Raycast installed. |
| user-invokable | true |
| args | [] |
A Raycast script that lets you paste clipboard images into a terminal-based AI agent as proper file attachments.
Terminal-based AI agents can't receive raw image data from the clipboard — pressing Cmd+V just types a "v". This script bridges the gap.
~/Pictures/clip-img/Copy scripts/clip-img.sh to ~/raycast-scripts/ and make it executable:
mkdir -p ~/raycast-scripts
cp scripts/clip-img.sh ~/raycast-scripts/
chmod +x ~/raycast-scripts/clip-img.sh
Open Raycast Settings → Extensions → Script Commands. Click Add Directories and select ~/raycast-scripts.
The "Paste Image to Agent" command will appear immediately under Script Commands.
Click the command row and set a hotkey in the Hotkey column (e.g. ⌥⇧V).
The script uses System Events to simulate Cmd+V, which requires Accessibility access.
Go to System Settings → Privacy & Security → Accessibility and ensure Raycast is enabled.
Add the image directory to your agent's allowed read paths. For example, in your agent's settings, allow reads from ~/Pictures/clip-img/*.
⌥⇧V)scripts/clip-img.sh — the Raycast script~/Pictures/clip-img/ (override with CLIP_IMG_DIR env var)