Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed PDF translation helper that sends PDF content to a user-configured translation endpoint and uses local PDF/OCR tools for that purpose.
Install only if you are comfortable sending PDF text to the OpenAI-compatible endpoint you configure and with a locally installed pdf2zh-next engine receiving your API key. Avoid confidential medical, legal, unpublished, or regulated documents unless that endpoint and engine are approved for that data.
export DOC_HOLMES_OPENAI_API_KEY=<your key> # created by you on that platform ``` SiliconFlow (`https://api.siliconflow.cn/v1`) or any OpenAI-compatible endpoint works the same way. > ⚠️ About GLM **Coding Plan** subscription endpoints (`/api/coding/paas/v4`): per the official FAQ, the plan only covers designated coding tools. Calls from other tools do NOT consume the plan quota - they are billed per-token against your account balance, and accounts shared across multiple people may face subscription restrictions. If you still want to connect your subscription, set `DOC_HOLMES_ALLOW_CODING_ENDPOINT=1` (one-time confirmation).
gpu = "未检测到"
try:
import subprocess as sp
r = sp.run(["nvidia-smi", "-L"], capture_output=True, text=True, timeout=10)
if r.returncode == 0 and r.stdout.strip():
gpu = r.stdout.strip().splitlines()[0]
except Exception:"OpenAI 兼容端点 + key（三选一）：\n"
" A) 智谱开放平台官方 API（推荐）：https://open.bigmodel.cn/api/paas/v4\n"
" glm-4.5-flash 免费档；在 open.bigmodel.cn 注册并创建你自己的 key\n"
" B) SiliconFlow 等聚合平台免费档（https://api.siliconflow.cn/v1）\n"
" C) 任意 OpenAI 兼容端点\n"
"设置方式（二选一）：\n"
" 1) 环境变量：\n"" 1) 环境变量：export DOC_HOLMES_OPENAI_API_KEY=<你的key>\n"
" 2) 配置文件：%s\n"
" 写入一行：DOC_HOLMES_OPENAI_API_KEY=<你的key>\n"
" （建议 chmod 600，key 明文仅本人可读）" % (cfg.base_url, env_file)
)
return cfgtimed_out = False
window = t0 - 1 # 产物发现窗口：只认本次运行开始后的文件
try:
proc = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout_s)
rc = proc.returncode
err_tail = ((proc.stderr or "") + (proc.stdout or ""))[-2000:]
except subprocess.TimeoutExpired as exc:return info
info["binary"] = binary
try:
out = subprocess.run([binary, "--version"], capture_output=True, text=True,
timeout=15)
info["version"] = (out.stdout or out.stderr).splitlines()[0].strip()
except Exception as exc:return info
info["binary"] = binary
try:
out = subprocess.run([binary, "--version"], capture_output=True, text=True,
timeout=15)
info["version"] = (out.stdout or out.stderr).splitlines()[0].strip()
except Exception as exc:info["reason"] = "tesseract 无法执行：%s" % exc
return info
try:
langs = subprocess.run([binary, "--list-langs"], capture_output=True,
text=True, timeout=15).stdout
info["langs"] = [x.strip() for x in langs.splitlines()[1:] if x.strip()]
except Exception:info["reason"] = "tesseract 无法执行：%s" % exc
return info
try:
langs = subprocess.run([binary, "--list-langs"], capture_output=True,
text=True, timeout=15).stdout
info["langs"] = [x.strip() for x in langs.splitlines()[1:] if x.strip()]
except Exception:page_pdf = base + ".pdf"
try:
# 单次调用同时产出 pdf + txt（tesseract 支持多输出配置，省一半 OCR 时间）
subprocess.run(
[info["binary"], img_path, base, "-l", lang, "--psm", psm,
"pdf", "txt"],
capture_output=True, timeout=page_timeout, check=True)No suspicious patterns detected.