diff --git a/src/interface.jl b/src/interface.jl
index 05d6842..880cb20 100755
--- a/src/interface.jl
+++ b/src/interface.jl
@@ -457,8 +457,8 @@ function conversation(a::agentReflex, usermsg::String; attemptlimit::Int=3)
   _ = addNewMessage(a, "user", usermsg)
   isusetools = isUseTools(a, usermsg)
   newinfo = extractinfo(a, usermsg)
-  a.env = newinfo !== nothing ? updateEnvState(a, newinfo)
-
+  a.env = newinfo !== nothing ? updateEnvState(a, newinfo) : a.env
+  error(1)
   #WORKING
   if isusetools # use tools before responseing
     response = work(a, usermsg)
@@ -485,21 +485,11 @@ end
 function work(a::agentReflex, usermsg::String)
   response = nothing
 
-  if a.thinkingmode == :new_thinking
-    _ = addNewMessage(a, "user", usermsg)
-    a.memory[:shortterm] = OrderedDict{String, Any}()
-    a.memory[:log] = OrderedDict{String, Any}()
-    a.memory[:shortterm]["user:"] = usermsg
-    a.memory[:log]["user:"] = usermsg
-    a.newplan = true
-  elseif a.thinkingmode == :continue_thinking
-    println("continue_thinking!!")
-    _ = addNewMessage(a, "user", usermsg)
-    a.memory[:shortterm]["Obs $(a.step):"] = usermsg
-    a.memory[:log]["Obs $(a.step):"] = usermsg
-  else
-    error("undefined condition thinkingmode = $thinkingmode $(@__LINE__)")
-  end
+  a.memory[:shortterm] = OrderedDict{String, Any}()
+  a.memory[:log] = OrderedDict{String, Any}()
+  a.memory[:shortterm]["user:"] = usermsg
+  a.memory[:log]["user:"] = usermsg
+  a.newplan = true
 
   while true  # Work loop
     # plan
@@ -600,6 +590,124 @@ function work(a::agentReflex, usermsg::String)
   return  response
 end
 
+# function work(a::agentReflex, usermsg::String)
+#   response = nothing
+
+#   if a.thinkingmode == :new_thinking
+#     _ = addNewMessage(a, "user", usermsg)
+#     a.memory[:shortterm] = OrderedDict{String, Any}()
+#     a.memory[:log] = OrderedDict{String, Any}()
+#     a.memory[:shortterm]["user:"] = usermsg
+#     a.memory[:log]["user:"] = usermsg
+#     a.newplan = true
+#   elseif a.thinkingmode == :continue_thinking
+#     println("continue_thinking!!")
+#     _ = addNewMessage(a, "user", usermsg)
+#     a.memory[:shortterm]["Obs $(a.step):"] = usermsg
+#     a.memory[:log]["Obs $(a.step):"] = usermsg
+#   else
+#     error("undefined condition thinkingmode = $thinkingmode $(@__LINE__)")
+#   end
+
+#   while true  # Work loop
+#     # plan
+#     if a.attempt <= a.attemptlimit
+#       toolname = nothing
+#       toolinput = nothing
+#       if a.newplan == true
+#         a.attempt += 1
+#         a.step = 0
+#         prompt_plan = planner_mistral_openorca(a)
+#         println("")
+#         @show prompt_plan
+#         response = sendReceivePrompt(a, prompt_plan, max_tokens=1024)
+
+#         # sometimes LLM add not-need word I don't want
+#         plan = splittext(response, ["Step 1", "<|im_end|>", "Response", "Execution", 
+#                                   "Result", "Recommendation", "My response"])
+#         # plan = replace(plan, "Plan:"=>"")
+#         println("")
+#         @show plan
+        
+#         a.newplan = false
+#         a.memory[:shortterm]["Plan $(a.attempt):"] = plan
+#         a.memory[:log]["Plan $(a.attempt):"] = plan
+#       end
+      
+#       println("")
+#       @show a.attempt
+
+#       # enter actor loop
+#       actorstate, msgToUser = actor(a)
+
+#       if actorstate == "chatbox"
+#         response = msgToUser
+#         break
+#       elseif actorstate == "all steps done" ||  actorstate == "formulateUserresponse"
+#         println("all steps done")
+
+#         response = formulateUserresponse(a)
+
+#         println("")
+#         formulatedresponse = response
+#         @show formulatedresponse
+        
+#         a.memory[:shortterm]["response $(a.attempt):"] = response
+#         a.memory[:log]["response $(a.attempt):"] = response
+
+#         # evaluate. if score < 8/10 try again.
+#         guideline = writeEvaluationGuideline(a, a.memory[:shortterm]["user:"])
+
+#         println("")
+#         @show guideline
+
+#         score = grading(a, guideline, response)
+#         @show score
+#         if score >= 6 # good enough answer
+#           break
+#         else  # self evaluate and reflect then try again
+#           analysis = analyze(a)
+#           println("")
+#           @show analysis
+          
+#           lessonwithcontext = selfReflext(a, analysis)
+          
+#           println("")
+#           @show lessonwithcontext
+
+#           newdict = OrderedDict()
+
+#           a.memory[:shortterm] = keepOnlyKeys(a.memory[:shortterm], ["user:"])
+
+#           headerToDetect = ["Lesson:", "Context:", ]
+#           headers = detectCharacters(lessonwithcontext, headerToDetect)
+#           chunkedtext = chunktext(lessonwithcontext, headers)
+
+#           a.memory[:longterm][chunkedtext["Context:"]] = chunkedtext["Lesson:"]
+#           a.newplan = true
+#           println("")
+#           println("RETRY $(a.attempt +1)")
+#           println("")
+          
+#         end
+#       else
+#         error("undefied condition, actorstate $actorstate $(@__LINE__)")
+#         break
+#       end
+#     else
+#       error("attempt limit reach")
+#       break
+#     end
+#   end
+
+#   # good enough answer
+
+
+#   # communicates with user
+#   _ = addNewMessage(a, "assistant", response)
+#   return  response
+# end
+
 
 # function evaluate()
   
diff --git a/src/type.jl b/src/type.jl
index 4fa1e74..4c74276 100644
--- a/src/type.jl
+++ b/src/type.jl
@@ -165,7 +165,7 @@ function agentReflex(
       p.s.1 each step should be a single action.
       p.s.2 don't respond to the stimulus yet.
       """,
-      :actor=> 
+    :actor=> 
       """
       Use the following format:
       Thought: you should always think about do you have all the required info and what to do according to step {step} of the plan and the info you have (pay attention to correct numeral calculation and commonsense).
diff --git a/src/utils.jl b/src/utils.jl
index 4eae588..738e0e6 100644
--- a/src/utils.jl
+++ b/src/utils.jl
@@ -363,40 +363,83 @@ end
 #   return thinkingmode 
 # end
 
+# function isUseTools(a::agentReflex, usermsg::String)
+#   prompt =
+#     """
+#     <|im_start|>system
+#     {systemMsg}
+
+#     Your earlier conversation with the user:
+#     None
+
+#     User's message:
+#     $usermsg
+
+#     From the user's message, Are there any explicit request from the user? Answer: {Yes/No/Not sure}. What do you need to do?
+#     <|im_end|>
+#     <|im_start|>assistant
+#     Answer: 
+#     """
+#     toollines = ""
+#     for (toolname, v) in a.tools
+#       if toolname ∉ [:chatbox]
+#         toolline = "$toolname: $(v[:description]) $(v[:input]) $(v[:output])\n"
+#         toollines *= toolline
+#       end
+#     end
+#     prompt = replace(prompt, "{systemMsg}" => a.roles[a.role])
+#     prompt = replace(prompt, "{tools}" => toollines)
+    
+#     result = sendReceivePrompt(a, prompt, temperature=0.2)
+
+#     if occursin("Yes", result)
+#       isUseTool = result
+#       @show isUseTool
+#       return true
+#     else
+#       return false
+#     end
+# end
+
 function isUseTools(a::agentReflex, usermsg::String)
-  prompt =
-    """
-    <|im_start|>system
-    {systemMsg}
-    You have access to the following tools:
-    {tools}
-
-    User's message:
-    {input}
-
-    From the user's message, Do you need to any tools before responseing? Answer: {Yes/No/Not sure}. What will the you do?
-    <|im_end|>
-    <|im_start|>assistant
-    Answer: 
-    """
     toollines = ""
     for (toolname, v) in a.tools
-      if toolname ∉ [:chatbox]
+      if toolname ∉ ["chatbox"]
         toolline = "$toolname: $(v[:description]) $(v[:input]) $(v[:output])\n"
         toollines *= toolline
       end
     end
-    prompt = replace(prompt, "{systemMsg}" => a.roles[a.role])
-    prompt = replace(prompt, "{tools}" => toollines)
-    prompt = replace(prompt, "{input}" => usermsg)
-    
-    result = sendReceivePrompt(a, prompt, temperature=0.2)
 
-    if occursin("Yes", result)
-      return true
-    else
-      return false
+    prompt =
+    """
+    <|im_start|>system
+    $(a.roles[a.role])
+    You have access to the following tools:
+    $toollines
+
+    Your earlier conversation with the user:
+    None
+
+    User's message:
+    $usermsg
+
+    From the user's message, what to do?
+    <|im_end|>
+    <|im_start|>assistant
+
+    """
+    
+    # if LLM mentions any tools, use Plan/Thought/Act loop
+    isusetool = false
+    result = sendReceivePrompt(a, prompt, temperature=0.2)
+    for (toolname, v) in a.tools
+      if occursin(toolname, result)
+        isusetool = true
+        break
+      end
     end
+
+    return isusetool
 end
 
 
@@ -441,7 +484,7 @@ function conversationSummary(a::T) where {T<:agent}
 
   """
   conversation = ""
-  summary = "nothing"
+  summary = ""
   if length(a.messages)!= 0
     for msg in a.messages[1:end-1]
       role = msg[:role]
@@ -458,7 +501,7 @@ function conversationSummary(a::T) where {T<:agent}
 
     prompt = replace(prompt, "{conversation}" => conversation)
     result = sendReceivePrompt(a, prompt)
-    summary = result === nothing ? "nothing" : result
+    summary = result === nothing ? "" : result
     summary = split(summary, "<|im_end|>")[1]
     if summary[1:1] == "\n"
       summary = summary[2:end]