diff --git a/Markdown Search.alfredworkflow b/Markdown Search.alfredworkflow
deleted file mode 100644
index 02ea229..0000000
Binary files a/Markdown Search.alfredworkflow and /dev/null differ
diff --git a/README.md b/README.md
index 16301a7..2167478 100644
--- a/README.md
+++ b/README.md
@@ -1,20 +1,67 @@
 # alfred-markdown-search
-Alfred Workflow for search and open markdown files
 
+Alfred Workflow for search and open Markdown files
 
-## Download
+## Documents
 
-Coming Soon...
+* [English document](REAMDME.md)
+* [中文文档](README_ZH.md)
 
 
-## Params
+## Pain Point
 
-* `MARKDOWN_PATH`
+When using Typora to manage markdown documents, the built-in search function is not very useful. Some documents cannot be searched.
 
 
+## Params Setting
+
+Open the environment variable setting page of `Markdown Search workflow`, and configure:
+
+* `MARKDOWN_PATH` -- `Required` -- Directory path of markdown files,multiple directories are separated by English colon `:`
+* `MARKDOWN_APP` -- `Optional` -- Third-party apps, such as `/Applications/Typora.app`
+
 ## How to use
 
-Keyword: `mds`
+Keyword: `mdd` (you can also customize this shortcut to trigger workflow)
+
+
+1. When no parameters are entered, 20 document records are displayed in reverse order according to the latest modification time;
+2. Support the `-t` (`title`) parameter to search the title area of the document (`title area`: to support multiple md document formats, define the `first nine lines` of the document as the title area for keyword search) ；
+3. Support inputting multiple keywords to filter, separated by spaces, keywords are not case sensitive;
+4. Select the corresponding line and press `Enter` to open the document with the system default markdown document App;
+5. Select the corresponding line and press `Options + Enter`, the document will be opened with a custom-configured third-party App;
+
+
+## Examples
+
+```
+# By default, 20 document records are displayed according to the latest modification time
+mdd
+
+
+# Search keywords in the title area, the keywords are not case sensitive
+mdd -t golang
+
+
+# Full text search keywords, keywords are not case sensitive
+mdd golang
+
+
+# Full text search keywords, multiple keywords are separated by spaces, keywords are not case sensitive
+mdd golang Python
+
+
+# Find documents that contain `golang` in the title area and `python` in the content of the document
+mdd -t golang python
+
+
+# Find documents with `golang` and `python` in the title area
+mdd -t golang,python
+
+
+# Find documents that contain `golang` and `python` keywords in the title area and `mysql` in the document content
+mdd -t golang,python mysql
+```
 
 
 ## Thanks
diff --git a/README_ZH.md b/README_ZH.md
new file mode 100644
index 0000000..d174d85
--- /dev/null
+++ b/README_ZH.md
@@ -0,0 +1,68 @@
+# alfred-markdown-search
+
+通过 Alfred Workflow 搜索和打开 Markdown 文档
+
+## 文档
+
+* [English document](REAMDME.md)
+* [中文文档](README_ZH.md)
+
+
+## 痛点
+
+在使用 Typora 管理 Markdown 文档时，自带的搜索功能不是很好用，有些文档搜索不到。
+
+
+## 参数设置
+
+打开 `Markdown Search workflow` 的环境变量设置页面，配置如下环境变量：
+
+* `MARKDOWN_PATH` -- `必填项` -- markdown文档所在目录路径，多个目录用英文冒号`:`分隔
+* `MARKDOWN_APP` -- `可选项` -- 第三方App, 例如 `/Applications/Typora.app`
+
+## 如何使用
+
+关键词: `mdd` （也可以自定义该快捷键触发 workflow）
+
+1. 不输入任何参数时，按照最近修改时间倒序展示20条文档记录；
+2. 支持 `-t` (`title`)参数，对文档标题区域搜索（`标题区域`：为支持多种md文档格式，定义文档的 `前九行` 为标题区域来进行关键字查找）；
+3. 支持输入多个关键字过滤，中间以空格分隔，关键字不区分大小写；
+4. 选中对应行，`回车`，会用系统默认的markdown文档App打开该文档；
+5. 选中对应行，按下 `Options + 回车`，会用自定义配置的第三方App打开该文档；
+
+
+## 示例
+
+```
+# 不输入任何参数，按最近修改时间展示20条文档记录
+mdd
+
+
+# 在标题区域搜索关键词，关键词不区分大小写
+mdd -t golang
+
+
+# 全文搜索关键词，关键词不区分大小写
+mdd golang
+
+
+# 全文搜索关键词，多个关键词之间用空格分隔，关键词不区分大小写
+mdd golang Python
+
+
+# 查找标题区域含有 golang ，且文档内容中包含 python 的文档
+mdd -t golang python
+
+
+# 查找标题区域含有 golang 和 python 的文档
+mdd -t golang,python
+
+
+# 查找标题区域含有 golang 和 python 关键词，且文档内容中包含 mysql 的文档
+mdd -t golang,python mysql
+```
+
+
+## 感谢
+
+* [MWeb workflow](https://github.com/tianhao/alfred-mweb-workflow)
diff --git a/shell/md-search.sh b/shell/md-search.sh
index 687bbfe..f051fa5 100755
--- a/shell/md-search.sh
+++ b/shell/md-search.sh
@@ -92,18 +92,26 @@ fi
 final_expr="${final_expr} | head -20" # 取前20个
 
 
-echo "{\"items\":["
+result="{\"items\":["
 n=0
+r=0
 while read line
 do
 #    echo line=${line}
 
     # 将文件名作为结果列表中的标题显示
     h="${line##*/}"
-    echo "{\"type\": \"file\",\"title\": \"${h}\",\"subtitle\": \"${line}\",\"arg\": \"${line}\"},"
+    result+="{\"type\": \"file\",\"title\": \"${h}\",\"subtitle\": \"${line}\",\"arg\": \"${line}\"},"
     n=$((n+1))
 done < <(eval "${final_expr}")
 if [ ${n} -eq 0 ]; then
-    echo "{\"title\": \"没有找到符合条件的文档\"}"
+    result+="{\"title\": \"没有找到符合条件的文档\"}"
 fi
-echo "]}"
\ No newline at end of file
+
+# Remove the json str last comma
+if [ ${r} -eq 0 ]; then
+    result=${result%,}
+fi
+
+result+="]}"
+echo ${result}
\ No newline at end of file
diff --git a/shell/test.sh b/shell/test.sh
index eb4ddc6..e9e6dc5 100755
--- a/shell/test.sh
+++ b/shell/test.sh
@@ -1,15 +1,20 @@
 #!/usr/bin/env bash
 
-# MARKDOWN_PATH="${HOME}/demo"
-# NOTE: 
+# debug mode true or false
+debug=false
+
+if [ "$debug" = true ]; then
+    MARKDOWN_PATH="${HOME}/demo/mnk"
+fi
 
 if [ -z "${MARKDOWN_PATH}" ];then
    echo "{ \"items\":[{\"type\": \"error\",\"title\": \"请先设置文档路径 MARKDOWN_PATH \"}]}"
    exit 1
 fi
 
-# echo "${MARKDOWN_PATH}"
-
+if [ "$debug" = true ]; then
+    echo "13--${MARKDOWN_PATH}"
+fi
 
 declare -a header_arr # 标题数组
 declare -a keyword_arr # 关键字数组
@@ -48,15 +53,19 @@ get_params $*
 final_expr=""
 IFS=":";for i in ${MARKDOWN_PATH}
 do
-    final_expr="${final_expr} find \"${i}\" -type f -iname '*.md';"
+    final_expr="${final_expr} find \"${i}\" -type f -iname '*.md' -print0;"
 done
 unset IFS
 final_expr="{${final_expr}}"
 #final_expr="{${final_expr}} | xargs -I'{}' stat -f'%m %N' '{}' | sort -rn | cut -d' ' -f 2-"  # | xargs -I'{}' grep -l tomcat '{}"
 
-# echo "${final_expr}"
-# echo "header_arr:${header_arr} -- keyword_arr:${keyword_arr} -- next_input:${next_input} -- last_input:${last_input} -- end_char:${end_char}"
+if [ "$debug" = true ]; then
+    echo "60--${final_expr}"
+    echo "header_arr:[${header_arr}] -- keyword_arr:[${keyword_arr}] -- next_input:[${next_input}] -- last_input:[${last_input}] -- end_char:[${end_char}]"
+fi
 
+# 按时间排序
+final_expr="${final_expr} | xargs -0 ls -t"
 
 # 如果有输入-h参数，过滤文档标题
 # 思路: grep -n 会输出行号，找到符合所有关键字的行，筛选行号=1的文档就可以了
@@ -70,9 +79,9 @@ if [ ${#header_arr[@]} -gt 0 ];then
     final_expr="${final_expr} | egrep '^.+\.md\:[1-9]\:' | awk -F':' '{print \$1}'"
 fi
 
-# echo "${final_expr}"
-# ➜ { find "/demo" -type f -iname '*.md';} | xargs -I'{}' grep -inHe 'aaa' '{}' | egrep '^.+\.md\:[1-9]\:' | awk -F':' '{print $1}'
-
+if [ "$debug" = true ]; then
+    echo "78--${final_expr}"
+fi
 
 # 如果有输入关键字，则用关键字筛选文档，并且按照文档标题匹配度排序
 if [ ${#keyword_arr[@]} -gt 0 ];then
@@ -81,11 +90,13 @@ if [ ${#keyword_arr[@]} -gt 0 ];then
         final_expr="${final_expr} | xargs -I'{}' grep -ile '${i}' '{}' | awk -F':' '{print \$1}' | uniq "
     done
 
-    # echo "83--${final_expr}"
+    if [ "$debug" = true ]; then
+        echo "89--${final_expr}"
+    fi
 
     # 按时间排序
     # final_expr="${final_expr} | xargs -I'{}' stat -f'%m %N' '{}' | sort -rn | cut -d' ' -f 2- "
-    final_expr="${final_expr} | xargs -I'{}' ls -t '{}'"
+    # final_expr="${final_expr} | xargs -I'{}' ls -t '{}'"
 
     # 排序表达式: 统计第一行匹配关键字个数，将匹配个数大的放在前面
     # 第一步: 输入"文件名",输出"文件名 匹配个数"
@@ -95,32 +106,46 @@ if [ ${#keyword_arr[@]} -gt 0 ];then
     egrep_expr=$(echo "${keyword_arr[@]}" | sed "s/[[:blank:]]/|/g")
     sort_expr="awk -F'\\n' '{system(\"egrep -ioe \\\"${egrep_expr}\\\" <<< \`head -1 \\\"\"\$1 \"\\\"\` | wc -l | xargs -I\\\"{}\\\" echo \\\"{}\\\" \\\"\"\$1\"\\\"\")}' | sort -r | cut -d' ' -f 2-"
     final_expr="${final_expr} | ${sort_expr} "
+
+    if [ "$debug" = true ]; then
+        echo "106--${final_expr}"
+    fi
 else
     # 按时间排序
     # final_expr="${final_expr} | xargs -I'{}' stat -f'%m %N' '{}' | sort -rn | cut -d' ' -f 2- "
-    final_expr="${final_expr} | xargs -I'{}' ls -t '{}'"
+    # final_expr="${final_expr} | xargs -0 ls -t"
+
+    if [ "$debug" = true ]; then
+        echo "114--${final_expr}"
+    fi
 fi
 
 final_expr="${final_expr} | head -20" # 取前20个
 
+if [ "$debug" = true ]; then
+    echo "121--${final_expr}"
+fi
 
-# echo "104--${final_expr}"
-
-
-echo "{\"items\":["
+result="{\"items\":["
 n=0
+r=0
 while read line
 do
 #    echo line=${line}
 
-    # 将文件中的前两行作为结果列表中的标题显示
-    # h="$(head -2 "${line}"| sed 's/-//g;s/\\/\\\\/g;s/"/\\"/g;s/[[:space:]]*$//g')"
     # 将文件名作为结果列表中的标题显示
     h="${line##*/}"
-    echo "{\"type\": \"file\",\"title\": \"${h}\",\"subtitle\": \"${line}\",\"arg\": \"${line}\"},"
+    result+="{\"type\": \"file\",\"title\": \"${h}\",\"subtitle\": \"${line}\",\"arg\": \"${line}\"},"
     n=$((n+1))
 done < <(eval "${final_expr}")
 if [ ${n} -eq 0 ]; then
-    echo "{\"title\": \"没有找到符合条件的文档\"}"
+    result+="{\"title\": \"没有找到符合条件的文档\"}"
 fi
-echo "]}"
\ No newline at end of file
+
+# Remove the json str last comma
+if [ ${r} -eq 0 ]; then
+    result=${result%,}
+fi
+
+result+="]}"
+echo ${result}
\ No newline at end of file