Wfgy Promptbundle Tools¶
Configuration File: wfgy_promptbundle_tools.json
Tool Type: Local
Tools Count: 1
This page contains all tools defined in the wfgy_promptbundle_tools.json configuration file.
Available Tools¶
WFGY_triage_llm_rag_failure (Type: WFGYPromptBundleTool)¶
Generate a structured prompt bundle (system + user message) to triage LLM/RAG failures using the …
WFGY_triage_llm_rag_failure tool specification
Tool Information:
Name:
WFGY_triage_llm_rag_failureType:
WFGYPromptBundleToolDescription: Generate a structured prompt bundle (system + user message) to triage LLM/RAG failures using the WFGY ProblemMap (No.1..No.16). Returns prompts, a minimal-fix checklist, and links to the public ProblemMap. No LLM or external API call is made.
Parameters:
bug_description(string) (required) Free-text description of the LLM/RAG incident. Include the user prompt, retrieved context, model answer, and any relevant logs.audience(string) (optional) Target audience for the prompt bundle tone. ‘beginner’ uses plain language, ‘engineer’ is concise and diagnostic, ‘infra’ is ops-focused.
Example Usage:
query = {
"name": "WFGY_triage_llm_rag_failure",
"arguments": {
"bug_description": "example_value"
}
}
result = tu.run(query)