# Add to your Claude Code skills
git clone https://github.com/usepr/evaLast scanned: 7/22/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-07-22T06:26:59.870Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}eva is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by usepr. 单文件智能体. It has 400 GitHub stars.
Yes. eva passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/usepr/eva" and add it to your Claude Code skills directory (see the Installation section above).
eva is primarily written in Python. It is open-source under usepr on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh eva against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
EVA是个麻雀虽小、五脏俱全的Agent智能体,相当于低配版Claude Code,能帮你写脚本、写测试案例、执行shell、分析数据等。我自己就是EVA的重度用户,日常处理各种任务。
各种好玩案例见当前仓库的 showcase 🦖🦖🦖
eva.py,有python就能运行eva -au '计算100w以内所有素数和并写到/tmp/result.txt'。当前就借助-asu选项将EVA接入了微信Bot直接创建一个eva.py并复制本仓库的eva.py文本内容粘贴进去(docker环境、运维环境等也很容易粘贴代码,无需复杂安装,Just Paste and Go)。当然,你也可以git clone本仓库。
在终端执行export EVA_API_KEY=你的deepseek API key(Windows系统则是set命令)
EVA支持OpenAI接口形式的LLM,可以是Ollma、vLLM拉起的本地模型,也可以是DeepSeek、OpenAI等官网API。切换方法是设置EVA_BASE_URL, EVA_MODEL_NAME, EVA_API_KEY这三个环境变量:
Linux设置方法:
export EVA_BASE_URL=http://xxxxxxxxx/v1
export EVA_MODEL_NAME=xxxxx
export EVA_API_KEY=sk-xxxxx
macOS 设置方法(zsh,如需在 macOS 上长期生效,可以将上述 export 配置写入~/.zshrc):
export EVA_BASE_URL=http://xxxxxxxxx/v1
export EVA_MODEL_NAME=xxxxx
export EVA_API_KEY=sk-xxxxx
Windows 命令行设置方法:
set EVA_BASE_URL=http://xxxxxxxxx/v1
set EVA_MODEL_NAME=xxxxx
set EVA_API_KEY=sk-xxxxx
Windows PowerShell设置方法:
$env:EVA_BASE_URL="http://xxxxxxxxx/v1"
$env:EVA_MODEL_NAME="xxxxx"
$env:EVA_API_KEY="sk-xxxxx"
不想每次设置环境变量? 也可以直接把配置写入
eva.py:打开文件顶部「LLM配置区」,将EVA_BASE_URL、EVA_MODEL_NAME、EVA_API_KEY三行引号内的值改成你的配置即可,无需在终端执行export。默认使用 DeepSeek 官方接口,通常你只需要填写EVA_API_KEY一行,例如:EVA_API_KEY = os.environ.get("EVA_API_KEY") or "sk-xxxxx" # ← 在此填入你的 API Key若已通过上方环境变量方式配置,则以环境变量为准(文件内配置仅作兜底)。
python3 eva.py。首次运行会生成eva脚本,Linux 下执行source ~/.bashrc让脚本生效;macOS 下执行source ~/.zshrc让脚本生效。后续直接输入命令eva即可输入增强说明:
prompt_toolkit,EVA会自动开启多行输入:Enter提交,Ctrl+N换行;如果终端支持,也可以用Alt+Enter换行输出增强说明:
rich,EVA会自动开启rich美化输出eva支持的选项:
usage: eva.py [-h] [-a] [-l] [-c] [-u USER_ASK] [-s] [-g]
options:
-h, --help show this help message and exit
-a, --allow-all 允许所有命令无需用户确认即可执行
-l, --list-session 列出所有session
-c, --clear-session 清除当前目录session
-u USER_ASK, --user-ask USER_ASK
独立地针对一条用户提问执行EVA
-s, --with-session 搭配-u使用,载入并保存session
-g, --goal goal模式,循环直到达成目标
绝大部分同学都带上-a来启动eva,虽然很方便,但要对eva行为多加关注下。
1、EVA运行过程可以随时打断,无论是打断LLM推理、打断工具执行、还是退出EVA,都是按 Ctrl + C
2、打断是个很有用的行为,比如某个命令超时时间太久你不想再等待,或者你想起有个背景忘记向EVA澄清需要补充说明,或者你发现前面对话有错别字需要重新问。
3、打断后,可以继续输入新的问题,不需要重新启动EVA;如果想完全退出EVA,连续按两下Ctrl + C
.eva/EVA.md是EVA的设定,通过它你可以对EVA进行各种设定、甚至赋予它各种技能。除了手动编辑EVA.md,你可以在eva启动时让它"分析下xxx/xxx/skills目录中所有的文件,给我生成一份EVA.md",EVA会自动生成设定内容。