read-image

read-image is a skill for Claude Code, Codex from Jedeiah/codex-read-image. It costs 65 tokens per session (1,155 once invoked), scanned A, original, MIT.

A skill that turns image files or screenshots into a written description using a configured image-reading service.

In plain words
What is it for?
Use it to read screenshots, scanned documents, photos, charts, and visual error reports, then ask questions about the extracted content.
Why use it?
Text-based agents cannot reliably inspect visual details on their own. This helps extract error messages, chart information, document text, or scene descriptions from images.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/jedeiah/codex-read-image/read-image
Any agent
npx skills add Jedeiah/codex-read-image --skill read-image
Clone the repo
git clone --depth 1 https://github.com/Jedeiah/codex-read-image

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 read-image

README.md
[![agentmods](https://agentmods.dev/badge/skills/jedeiah/codex-read-image/read-image.svg)](https://agentmods.dev/skills/jedeiah/codex-read-image/read-image)
Your own site
<a href="https://agentmods.dev/skills/jedeiah/codex-read-image/read-image"><img src="https://agentmods.dev/badge/skills/jedeiah/codex-read-image/read-image.svg" alt="Measured on agentmods" height="20"></a>
Per session 65 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,155 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00065 $0.01155
Opus 5 $0.00032 $0.00577
Sonnet 5 $0.00013 $0.00231
Haiku 4.5 $0.00006 $0.00115

Measured 3d ago against content hash c5b45a3c04a0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

read-image 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 3d 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.

plugins/read-image/skills/read-image/SKILL.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Read Image(图片读取)

何时使用

  • 用户提供了图片路径或拖入图片,要求"看一下 / 读取 / 描述 / 分析 / 识别";
  • 需要把图片内容变成文字:截图里的报错、图表数据、设计稿、证件扫描件、照片等;
  • 用户提问时附带了图片文件。

工作流程

  1. 确认图片存在:检查用户给的路径是否真实存在,支持 png / jpg / jpeg / webp / gif。

  2. 运行脚本(脚本会先 base64 编码图片,再调用视觉模型)。脚本位于插件目录的 scripts/ 下,相对于本 SKILL.md(skills/read-image/)是 ../../scripts/read_image.py

    python3 <插件目录>/scripts/read_image.py <图片路径> [更多图片...] --prompt "<用户的具体问题>"
    

    查找脚本位置后运行,例如安装目录为 ~/.codex/.tmp/marketplaces/codex-read-image/plugins/read-image 时:

    python3 ~/.codex/.tmp/marketplaces/codex-read-image/plugins/read-image/scripts/read_image.py /tmp/screenshot.png \
      --prompt "详细描述这张截图的内容,包括其中的报错信息"
    
  3. 读取脚本输出:脚本把视觉模型的识别结果打印在终端,作为后续回答的事实依据。

  4. 回复用户:把识别结果转述给用户(中文),如果用户在原问题上继续追问,基于识别文本继续处理(总结、翻译、找 bug、提取数据等)。

配置检查(首次使用必须)

推荐方式:插件目录里的 .env(卸载插件时随插件一起删除)。首次使用如果只有 .env.example 模板,先复制一份:cp 插件目录/.env.example 插件目录/.env,再编辑填写。 也可以复制一份到 ~/.config/read-image/.env 作为备用(插件升级时不会被覆盖)。 两者结构相同,填好即可,不需要动 shell 配置:

环境变量 必填 说明
READ_IMAGE_API_KEY 视觉模型服务的 API 密钥
READ_IMAGE_BASE_URL OpenAI 兼容接口地址,默认 https://api.openai.com/v1
READ_IMAGE_MODEL 视觉模型名,默认 gpt-4o-mini
READ_IMAGE_THINKING 思考模式 auto/on/off,默认 auto(glm 系列自动开启,其他模型关闭)

常见可用的视觉模型示例:gpt-4o-mini / gpt-4o(OpenAI)、qwen-vl-max(通义千问,DashScope 兼容模式)、glm-4.6v-flash / glm-4v-flash(智谱,免费)、SiliconFlow 上各家开源视觉模型。

.env 示例(编辑插件目录里的 .env 填值):

READ_IMAGE_API_KEY=你的密钥
READ_IMAGE_MODEL=glm-4.6v-flash
READ_IMAGE_BASE_URL=https://open.bigmodel.cn/api/paas/v4
READ_IMAGE_THINKING=auto

上面是智谱免费模型示例(GLM-4.6V-Flash);不填时脚本默认使用 gpt-4o-mini + https://api.openai.com/v1。如果都没有 READ_IMAGE_API_KEY,脚本会明确报错。此时应告诉用户按上面示例补上,改完重启 Codex 再试。配置优先级:命令行参数 > 环境变量 > 插件目录 .env > ~/.config/read-image/.env > 内置默认值。.env 不在 Git 仓库里,插件升级不会覆盖它。

当前 DeepSeek 接口不支持图片输入,需要另配一个支持视觉的模型服务,在 .env 里填写即可。

注意事项

  • 大图片可能超过视觉模型限制:可先用 macOS 自带命令缩小(sips -Z 1024 原图 --out 缩小图.png)再识别。
  • 一次可以传多张图,脚本会逐张识别并分别输出。
  • 图片只会上传到配置的视觉模型服务;不要把 API 密钥写进 prompt 或输出。
  • .env 里存有你的密钥,不要把这个文件或整个插件目录分享给别人。
  • 识别结果来自第三方视觉模型,可能与原图存在细节差异,回答时注意这一点。

Read the full file on GitHub · 69 lines

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. 3d ago First seen · 69 lines · 65 tokens per session scan A c5b45a3c04a0

Subscribe to this mod's changes

read-image is a skill published in the GitHub repository Jedeiah/codex-read-image (3 stars, last pushed 1mo ago), licensed MIT. It adds 65 tokens to every session and 1,155 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

stop-that-shit

Keep coding agents focused on requested and necessary work. Use for bounded changes, review-only tasks, scope creep, speculative hardening, unnecessary hashing or dependencies, repeated audit loops, or when the user invokes Stop That Shit.

lennney/stop-that-shit · 50 tokens

codex-auth-setup

Install or refresh codex-multi-auth for the official Codex CLI, run first login, and verify account health and routing.

ndycode/codex-multi-auth · 32 tokens

example

Example fixture skill for scanner regression coverage.

ndycode/codex-multi-auth · 10 tokens

codexu-pr-review

Review codexU pull requests and decide whether they should be merged, changed, split, declined, or kept as a fork based on product positioning, roadmap, privacy, data semantics, native macOS design, architecture, scope, and verification. Use for PR review, mergeability assessment, contribution triage, and feature-fit…

shanggqm/codexU · 79 tokens

dev-flow

Use Dev Flow for bounded Codex software development tasks: implementation, bug fixes, refactoring, targeted testing, development delivery, and explicit parallel batches isolated in separate Git worktrees. It may be selected implicitly for those tasks or explicitly with $dev-flow-codex:dev-flow. Do not create a Dev…

Innocent-children/dev-flow · 84 tokens

codexkit-repository-maintenance

Use when maintaining or improving the GameStudio-CodexKIT source repository, including CI, governance, catalog, generators, adapters, packaging, documentation, versioning, or release readiness.

hoatv2211/GameStudio-CodexKIT · 45 tokens