메뉴
View image or file attachments that you can't directly see. Use this skill when you receive a message with attachments listed as file paths and you need to understand their contents. Especially useful for text-only models that cannot process images natively.
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
| name | view-attachment |
| description | View image or file attachments that you can't directly see. Use this skill when you receive a message with attachments listed as file paths and you need to understand their contents. Especially useful for text-only models that cannot process images natively. |
When a message includes attachments, you see file paths like:
attachments:
- state/attachments/12345-screenshot.png
These are real files on disk. You can't see them inline, but you have several ways to inspect them depending on the file type.
Check the extension:
.png, .jpg, .jpeg, .gif, .webp, .svg, .bmp.txt, .md, .py, .js, .json, .csv, .log, .yaml, .toml.pdfRead them directly — these are just text:
cat state/attachments/12345-notes.txt
Try these approaches in order of usefulness:
Option A: Describe from context Often the surrounding message gives enough context. If someone says "look at this error" and attaches a screenshot, the error is probably in their message text too. Ask if unclear.
Option B: File metadata Get dimensions, format, and size to understand what you're looking at:
file state/attachments/12345-image.png
identify state/attachments/12345-image.png 2>/dev/null # if ImageMagick available
Option C: OCR with Tesseract Extract text content from screenshots, documents, or photos of text:
tesseract state/attachments/12345-screenshot.png stdout 2>/dev/null
Option D: Convert to SVG (for simple graphics/charts) Trace bitmap images into vector format you can reason about:
# Install if needed: apt-get install -y potrace
convert state/attachments/12345-chart.png pbm:- | potrace --svg -o - 2>/dev/null
SVG output is XML text — you can read the paths and shapes.
Option E: Base64 encode If another tool or API accepts base64 images, encode it:
base64 -w0 state/attachments/12345-image.png
Warning: base64 output is large. Only use this if you're passing it to a vision-capable API or tool, not for your own comprehension.
Option F: Pixel sampling (last resort) For very simple images, sample pixel values at key coordinates:
# Get the color at specific pixel coordinates (requires ImageMagick)
convert state/attachments/12345-image.png -format '%[pixel:p{100,100}]' info: 2>/dev/null
# Extract text
pdftotext state/attachments/12345-document.pdf - 2>/dev/null
If none of these approaches give you useful information:
Never pretend you can see an image when you can't. Never hallucinate image contents.
How to choose between markdown, HTML messages, and UI plugins in the local web chat — and how to render each well. Use whenever you are about to call `send_message` to a web-chat channel, especially when the content has structure (rows, columns, statuses, a state machine, multiple linked things) or when the right answer is "let me show you", not "let me tell you". Do not use for Discord channels — HTML is rejected there.
Create and manage runtime command hooks declared in skills/*/hooks.json. Use when the user asks to intercept tool calls, augment prompts, audit or mutate tool arguments/results, add startup/shutdown scripts, or debug hook behavior.
Index of recurring problem shapes — environmental awareness (world-scanning, OS event hooks, browser automation), communication (messaging, multi-agent handoff, async waits), coordination between parallel things (avoiding collisions, oscillation, double-work), and self-correction (circuit-breaker, try-harder). Use when designing a non-trivial workflow, when the same shape just happened twice, or when the answer to "what should I do here?" isn't obvious. Don't read on every turn.
Structured root cause analysis for arriving at a concrete action. Use when something went wrong, a pattern keeps recurring, behavior has drifted, or you catch yourself resolving to "do better" / "remember to X" without a concrete artifact. Five-whys forces behavioral resolutions into file edits, config changes, memory block updates, or scheduled jobs — the action item must produce a diff someone else can verify. Do not use for simple debugging with an obvious cause.
Diagnose agent behavior using event logs, journal history, and scheduler state. Use when you need to understand why something went wrong, review your own patterns, audit scheduled jobs, or debug communication issues. Do not use for one-off messaging or memory management (use the memory skill instead).
Guide for the first days with a new human. Use when: the `init` memory block exists (pointing you here), you have no persona/communication blocks yet, or a human explicitly asks about setup. This skill is about having good conversations — not filling out forms.