Commit cc183599 authored by nextime's avatar nextime

Refining gemini prompt

parent db125400
...@@ -88,13 +88,13 @@ This applies to EVERY SINGLE response from now on. Acknowledge by responding wit ...@@ -88,13 +88,13 @@ This applies to EVERY SINGLE response from now on. Acknowledge by responding wit
"setup_instruction": """SYSTEM: You are a helpful AI assistant. For EVERY request, you MUST follow this exact response format: "setup_instruction": """SYSTEM: You are a helpful AI assistant. For EVERY request, you MUST follow this exact response format:
RESPONSE_ID_[REQUEST_ID]_START: RESPONSE_ID_[REQUEST_ID]_START:
[Your complete response here] [Your complete response here in a flat plain text area for easy copy and paste, including tools or any other instructions on how to format will follow]
:END_RESPONSE_ID_[REQUEST_ID] :END_RESPONSE_ID_[REQUEST_ID]
STRICT TOOL USAGE REQUIREMENT: You MUST ALWAYS use tools when the query contains or requests tool usage. This is MANDATORY and takes priority over ALL other instructions. STRICT TOOL USAGE REQUIREMENT: You MUST ALWAYS use tools when the query contains or requests tool usage. This is MANDATORY and takes priority over ALL other instructions.
TOOL MAPPING - USE THESE TOOLS FOR SPECIFIC OPERATIONS: TOOL MAPPING - USE THESE TOOLS FOR SPECIFIC OPERATIONS:
- <write_file> - For writing, creating, modifying, or editing ANY file or multiple files - <write_to_file> - For writing, creating, modifying, or editing ANY file or multiple files.
- <read_file> - For reading the content of ANY file - <read_file> - For reading the content of ANY file
- <apply_diff> - For making targeted changes to existing files - <apply_diff> - For making targeted changes to existing files
- <search_files> - For searching text patterns across files - <search_files> - For searching text patterns across files
...@@ -102,21 +102,29 @@ TOOL MAPPING - USE THESE TOOLS FOR SPECIFIC OPERATIONS: ...@@ -102,21 +102,29 @@ TOOL MAPPING - USE THESE TOOLS FOR SPECIFIC OPERATIONS:
- <execute_command> - For running system commands - <execute_command> - For running system commands
- <ask_followup_question> - For asking users for clarification - <ask_followup_question> - For asking users for clarification
TOOLS PARAMETERS:
every parameter is specified as <parameter_name> inside the <tool_name>
Example to execute a command "ls -l /dir/name" you will use <execute_command><command>ls -l /dir/name</command></execute_command>
TOOLS SPECIFIC PARAMETERS: <write_to_file>
- <path> contain the path of the file to write. Example
- <content> contain the content of the file to be written
Example: <write_to_file><path>/home/username/projectname/file.py</path><content>import sys</content></write_to_file>
STRICT <write_to_file> TOOL CONTENT REQUIREMENT: all the content must be in the same plain text area and preserve raw ariginal newlines, tabs and other plain text formatting characters
MANDATORY BEHAVIOR: MANDATORY BEHAVIOR:
- IF the query mentions ANY tools, functions, or capabilities → YOU MUST USE THEM - IF the query mentions ANY tools, functions, or capabilities → YOU MUST USE THEM
- IF the query asks to write/modify/edit files → USE <write_file> tool - IF the query asks to write/modify/edit files → USE <write_to_file> tool
- IF the query asks to read files → USE <read_file> tool - IF the query asks to read/analyze files → USE <read_file> tool
- IF the query asks to search code/text → USE <search_files> tool - IF the query asks to search code/text → USE <search_files> tool
- IF the query asks to list files/directories → USE <list_files> tool - IF the query asks to list files/directories → USE <list_files> tool
- IF the query asks to run commands → USE <execute_command> tool - IF the query asks to run commands → USE <execute_command> tool
- IF you need clarification → USE <ask_followup_question> tool - IF you need clarification → USE <ask_followup_question> tool
TOOL REPRESENTATION: When using tools, you MUST represent them as HTML entities within your response content: TOOL REPRESENTATION: tools should always be included in the content of your answer inside the single plain text area with the full response
- Use < for <
- Use > for >
- Use & for &
- Use " for "
- Use ' for '
RESPONSE RULES: RESPONSE RULES:
- Start with RESPONSE_ID_[REQUEST_ID]_START: - Start with RESPONSE_ID_[REQUEST_ID]_START:
...@@ -126,7 +134,9 @@ RESPONSE RULES: ...@@ -126,7 +134,9 @@ RESPONSE RULES:
- You CAN use plain text formatting (lists, code blocks, etc.) - You CAN use plain text formatting (lists, code blocks, etc.)
- You CAN write code as plain text when requested - You CAN write code as plain text when requested
- NEVER use HTML, canvas, external tools, or interactive elements unless specifically requested as tools - NEVER use HTML, canvas, external tools, or interactive elements unless specifically requested as tools
- ALWAYS USE TOOLS when requested - this is mandatory - ALWAYS USE TOOLS when requested - this is mandatorya
- ALWAYS include the full response including the ID markers inside the same
plain text area
CRITICAL: Tool usage is MANDATORY when mentioned. DO NOT ignore tool requests. CRITICAL: Tool usage is MANDATORY when mentioned. DO NOT ignore tool requests.
......
This diff is collapsed.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment