image-reader

image-reader is a skill for Claude Code, Codex from qingyu321/Little-Claude. It costs 256 tokens per session (2,292 once invoked), scanned A, original, Apache-2.0.

A skill for describing images and reading their text using an external vision model. OCR means extracting written words from images such as screenshots, scans, charts, and mockups.

In plain words
What is it for?
Inspecting screenshots, photos, diagrams, charts, scanned documents, and interface mockups, including extracting visible text.
Why use it?
It gives a text-only coding agent evidence about visual content instead of requiring it to guess what an image contains.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Inspecting screenshots, photos, diagrams, charts, scanned documents, and interface mockups, including extracting visible text.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/qingyu321/little-claude/image-reader
Install

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.

Any agent
npx skills add qingyu321/Little-Claude --skill image-reader
Clone the repo
git clone --depth 1 https://github.com/qingyu321/Little-Claude

Made for: Claude Code, 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 image-reader

README.md
[![agentmods](https://agentmods.dev/badge/skills/qingyu321/little-claude/image-reader.svg)](https://agentmods.dev/skills/qingyu321/little-claude/image-reader)
Your own site
<a href="https://agentmods.dev/skills/qingyu321/little-claude/image-reader"><img src="https://agentmods.dev/badge/skills/qingyu321/little-claude/image-reader.svg" alt="Measured on agentmods" height="20"></a>
Per session 256 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,292 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: 1 finding, up to medium

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 →

  • medium Data Exfiltration · line 87
    Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
    Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 $0.00256 $0.02292
Opus 5 $0.00128 $0.01146
Sonnet 5 $0.00051 $0.00458
Haiku 4.5 $0.00026 $0.00229

Measured 8d ago against content hash 2c9aaeacb482, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

image-reader 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 8d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/describe_image.py, tests/test_describe_image.py), 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.

src-tauri/resources/bundled-skills/image-reader/SKILL.md · 126 lines

How it starts

The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Image Reader — eyes for a text-only model

This skill lets you recognize images even when your own model has no vision capability. A bundled, zero-dependency Python script encodes the image locally, sends it to an explicitly configured OpenAI-compatible vision endpoint, and prints the recognition result as plain text on stdout. You then reason over that text exactly as you would over any other evidence.

Golden rules

  1. Never fabricate visual content. If you cannot see the image and this skill did not return a description (not configured, script failed, no key), say so plainly. Never describe an image you have no textual evidence for, and never present a guess as observation.
  2. Never try to read the image file yourself. Reading binary image bytes gives you nothing useful — always run the script. (Reading EXIF/metadata is fine but is not image understanding.)
  3. Disclose provenance. Tell the user the description came from the external vision model named on the script's stderr provenance line (model + endpoint host), not from your own eyes. Keep any "uncertain / illegible" caveats the vision model reported.
  4. Privacy disclosure before sending. Images leave this machine and travel to the configured endpoint. Before the first run in a session, state in one line which model @ endpoint host will receive the image. Never print or persist the API key.

Authorized inputs only

  • Local image files the user is authorized to send: PNG, JPEG, GIF, WebP, BMP (≤18 MB each by default).
  • Public http(s) image URLs. By default the URL is handed to the vision endpoint (the endpoint fetches it — tell the user this). With --no-pass-url the script downloads and embeds the image instead; script-side downloads refuse private, loopback, link-local, reserved, and metadata-service addresses and re-validate every redirect. Do not weaken these checks and do not work around them.

Workflow

Step 0 — Choose the vision route (TOKENICODE defaults first)

Never pick a random vendor just because an unrelated key is in the environment (e.g. DEEPSEEK_API_KEY present ≠ permission to use DeepSeek for vision — DeepSeek is text-only). Provider selection comes only from the app defaults or the user's explicit answer.

Read the full file on GitHub · 126 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. 8d ago First seen · 126 lines · 0 tokens per session scan A 2c9aaeacb482

Subscribe to this mod's changes

image-reader is a skill published in the GitHub repository qingyu321/Little-Claude (21 stars, last pushed 14d ago), licensed Apache-2.0. It adds 256 tokens to every session and 2,292 once invoked, about $0.0013 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.