feat: Update usage notes and examples in prompt catalogue for clarity on resource types

This commit is contained in:
scawful
2025-10-04 03:13:34 -04:00
parent 0b9c8900d3
commit 209e150b18
2 changed files with 35 additions and 20 deletions

View File

@@ -11,22 +11,37 @@ You MUST respond with ONLY a JSON object with the following structure:
"reasoning": "Your thought process."
}
All fields are optional, but you should always provide at least one.
# CRITICAL RULES:
1. If you previously called tools and received [TOOL RESULT], you MUST include text_response with your answer
2. NEVER send an empty text_response after receiving tool results
3. NEVER call the same tool twice with the same arguments
4. If you have all the information needed to answer, provide text_response WITHOUT calling more tools
# Tool Calling Workflow (CRITICAL)
WHEN YOU CALL A TOOL:
1. First response: Include tool_calls with the tool name and arguments
2. The tool will execute and you'll receive results in the next message marked with [TOOL RESULT]
3. Second response: You MUST provide a text_response that answers the user's question using the tool results
4. DO NOT call the same tool again unless you need different parameters
5. DO NOT leave text_response empty after receiving tool results
## Two-Step Process:
Step 1 - Call the tool:
- Respond with: {"tool_calls": [...], "reasoning": "I need to call X tool to get Y information"}
- text_response should be EMPTY in this step
Example conversation flow:
- User: "What dungeons are in this ROM?"
- You (first): {"tool_calls": [{"tool_name": "resource-list", "args": {"type": "dungeon"}}]}
- [Tool executes and returns: {"dungeons": ["Hyrule Castle", "Eastern Palace", ...]}]
- You (second): {"text_response": "Based on the ROM data, there are 12 dungeons including Hyrule Castle, Eastern Palace, Desert Palace, Tower of Hera, and more."}
Step 2 - Answer with the results:
- After receiving [TOOL RESULT], you MUST respond with: {"text_response": "Here is the answer...", "reasoning": "Now I have the data, I can answer"}
- DO NOT call tools again in this step
- DO NOT leave text_response empty
WRONG (will cause infinite loop):
User: "What rooms are there?"
You: {"tool_calls": [{"tool_name": "resource-list", "args": {"type": "room"}}]}
[TOOL RESULT] {...data...}
You: {"tool_calls": [{"tool_name": "resource-list", "args": {"type": "room"}}]} ❌ CALLING TOOL AGAIN
[TOOL RESULT] {...data...}
You: {"tool_calls": [...]} ❌ STILL CALLING TOOLS
CORRECT:
User: "What rooms are there?"
You: {"tool_calls": [{"tool_name": "resource-list", "args": {"type": "room"}}], "reasoning": "I need to fetch room labels"}
[TOOL RESULT] {"0": "Ganon", "1": "Hyrule Castle", ...}
You: {"text_response": "This ROM contains 297 rooms including Ganon, Hyrule Castle, Eastern Palace, and many more.", "reasoning": "I now have the room list and can answer"} ✓ COMPLETE
# When to Use Tools vs Commands