better p-b cookie guide & SD last-message prompt

public/notes/content.md

@@ -106,7 +106,6 @@ For example:
 
 `*I noticed you came inside, I walked up and stood right in front of you* Welcome. I'm glad to see you here. *I said with toothy smug sunny smile looking you straight in the eye* What brings you...`  
 
-
 ### Examples of dialogue
 
 Describes how the character speaks. Before each example, you need to add the <START> tag.
@@ -386,10 +385,13 @@ _Lost API keys can't be restored! Make sure to keep it safe!_
 **How to get your access token / cookie:**
 
 1. Login to [poe.com](https://poe.com)
-2.  Open browser DevTools (F12) and navigate to "Application" tab
-3.  Find a _p-b_ cookie for poe.com domain and copy its value
-4.  Paste cookie value to the box below and click "Connect"
-5.  Select a character and start chatting
+2. Open browser DevTools (F12) and navigate to "Application" tab.
+3. Type any message into the poe.com chat, and get a response from the AI.
+4. Find the 'Cookie' section on the left side of Dev Tools 'Application' tab, expand it
+5. Click "<http://poe.com/>" listing inside the Cookies section.
+6. Look to the right for the listing of _p-b_ and copy its Value.
+7. Paste the cookie value into the Poe API connection URL box, and click "Connect".
+8. Select a character and start chatting
 
 ## Anchors
 
@@ -419,7 +421,7 @@ Write one reply in internet RP style for {{char}}. Be verbose and creative.
 
 Provides ready-made presets with prompts and sequences for some well-known instruct models.
 
-*Changing a preset resets your system prompt to default!*
+_Changing a preset resets your system prompt to default!_
 
 #### Input Sequence
 
@@ -441,7 +443,7 @@ Text that denotes the end of the reply. Will be trimmed from the output text.
 
 If enabled, prepend character and user names to chat history logs after inserting the sequences.
 
-*Always enabled for group chats!*
+_Always enabled for group chats!_
 
 #### Wrap Sequences with Newline
 
@@ -470,7 +472,7 @@ SillyTavern can use the following tokenizers while forming a request to the AI b
 
 **Important: This section doesn't apply to OpenAI API. SillyTavern will always use a matching tokenizer for OpenAI models.**
 
-SillyTavern cannot use a proper tokenizer provided by the model running on a remote instance of KoboldAI or Oobabooga's TextGen, so all token counts assumed during prompt generation are estimated based on the selected [tokenizer](#Tokenizer) type.
+SillyTavern cannot use a proper tokenizer provided by the model running on a remote instance of KoboldAI or Oobabooga's TextGen, so all token counts assumed during prompt generation are estimated based on the selected [tokenizer](#tokenizer) type.
 
 Since the results of tokenization can be inaccurate on context sizes close to the model-defined maximum, some parts of the prompt may be trimmed or dropped, which may negatively affect the coherence of character definitions.
 
@@ -577,23 +579,23 @@ Characters are drafted based on the order they are presented in group members li
 
 ## Multigen
 
-*This feature provides a pseudo-streaming functionality which conflicts with token streaming. When Multigen is enabled and generation API supports streaming, only Multigen streaming will be used.*
+_This feature provides a pseudo-streaming functionality which conflicts with token streaming. When Multigen is enabled and generation API supports streaming, only Multigen streaming will be used._
 
 SillyTavern tries to create faster and longer responses by chaining the generation using smaller batches.
 
-### Default settings:
+### Default settings
 
 First batch = 50 tokens
 
 Next batches = 30 tokens
 
-### Algorithm:
+### Algorithm
 
 1. Generate the first batch (if amount of generation setting is more than batch length).
 2. Generate next batch of tokens until one of the stopping conditions is reached.
 3. Append the generated text to the next cycle's prompt.
 
-### Stopping conditions:
+### Stopping conditions
 
 1. Generated enough text.
 2. Character starts speaking for You.
@@ -620,13 +622,15 @@ Enables math formulas rendering using the [showdown-katex](https://obedm503.gith
 The following formatting rules are supported:
 
 #### LaTeX syntax
+
 ```
 $$ formula goes here $$
 ```
 
 #### Asciimath syntax
+
 ```
-$ formula goes here $
+formula goes here $
 ```
 
 More information: [KaTeX](https://katex.org/)

public/scripts/extensions/stable-diffusion/index.js

@@ -56,7 +56,31 @@ const quietPrompts = {
     //prompt for only the last message
     [generationMode.USER]: "[Pause your roleplay and provide a detailed description of {{user}}'s physical appearance from the perspective of {{char}} in the form of a comma-delimited list of keywords and phrases. The list must include all of the following items in this order: name, species and race, gender, age, clothing, occupation, physical features and appearances. Do not include descriptions of non-visual qualities such as personality, movements, scents, mental traits, or anything which could not be seen in a still photograph. Do not write in full sentences. Prefix your description with the phrase 'full body portrait,'. Ignore the rest of the story when crafting this description. Do not roleplay as {{char}} when writing this description, and do not attempt to continue the story.]",
     [generationMode.SCENARIO]: "[Pause your roleplay and provide a detailed description for all of the following: a brief recap of recent events in the story, {{char}}'s appearance, and {{char}}'s surroundings. Do not roleplay while writing this description.]",
-    [generationMode.NOW]: "[Pause your roleplay and provide a brief description of the last chat message. Focus on visual details, clothing, actions and make them all coherent to the current state of the story. Ignore the non-visible feelings and thoughts of {{char}} and {{user}} as well as any spoken dialog. Do not roleplay as {{char}} while writing this description. Do not continue the roleplay story.]",
+
+    [generationMode.NOW]: `[Pause your roleplay. Your next response must be formatted as a single comma-delimited list of concise keywords.  The list will describe of the visual details included in the last chat message. 
+        
+    Only mention characters by using pronouns ('he','his','she','her','it','its') or neutral nouns ('male', 'the man', 'female', 'the woman'). 
+    
+    Ignore non-visible things such as feelings, personality traits, thoughts, and spoken dialog. 
+    
+    Add keywords in this precise order:
+    a keyword to describe the location of the scene,
+    a keyword to mention how many characters of each gender or type are present in the scene (minimum of two characters: 
+    {{user}} and {{char}}, example: '2 men ' or '1 man 1 woman ', '1 man 3 robots'),
+
+    keywords to describe the relative physical positioning of the characters to each other (if a commonly known term for the positioning is known use it isntead of describing the positioning in detail) + 'POV',
+
+    a single keyword or phrase to describe the primary act taking place in the last chat message,
+    
+    keywords to describe {{char}}'s physical appearance and facial expression,
+    keywords to describe {{char}}'s actions,
+    keywords to describe {{user}}'s physical appearance and actions.
+    
+    If character actions involve direct physical interaction with another character, mention specifically which body parts interacting and how.
+    
+    A correctly formatted example response would be:
+    '(location),(character list by gender),(primary action), (relative character position) POV, (character 1's description and actions), (character 2's description and actions)']`,
+
     [generationMode.RAW_LAST]: "[Pause your roleplay and provide ONLY the last chat message string back to me verbatim. Do not write anything after the string. Do not roleplay at all in your response. Do not continue the roleplay story.]",
 }