Diagnose LLM output failures and generate corrected prompts with regression tests.
npx skills add https://github.com/vecear/Nipponverb --skill llm-debugger-vecear
Please help me install this Agent Skill. Skill: llm-debugger Source: https://github.com/vecear/Nipponverb/tree/main/.claude/skills/llm-debugger Command: npx skills add https://github.com/vecear/Nipponverb --skill llm-debugger-vecear
LLM outputs can be unreliable, containing hallucinations, format violations, and reasoning errors. This skill provides systematic diagnosis and fixes to improve reliability and trust in AI systems.
Run the LLM Debugger on a failing prompt to identify issues, generate a corrected prompt, and create regression tests.
High-intent search queries and answers about installing and using this skill.
Diagnose LLM hallucinations and reasoning errors by classifying the root cause of the failure. This skill provides prompt fixes, tool improvements, and regression tests to systematically resolve these issues and improve output reliability.
Stop LLM format violations and constraint breaches by identifying the specific failure type and applying targeted prompt fixes. The tool generates new test cases to verify that the corrected prompt enforces your required constraints.
The best way to debug LLM prompt failures is using a structured workflow that includes diagnosis, prompt refinement, test-case generation, and verification. This approach yields root-cause classification and iterative tool improvements for reliable AI quality.
Yes, you can automatically generate test cases for failing LLM workflows. The skill creates regression tests based on diagnosed failures, enabling iterative verification to ensure prompt fixes prevent future errors.
Identify the root cause of LLM tool misuse and incorrect outputs through systematic error diagnosis. The skill classifies failure types and recommends specific tool improvements to resolve the underlying reliability issues.
allexcdDiagnose, reproduce, and fix code defects from bug reports or error logs.
selcukyucelGenerate structured evaluation datasets with binary rubric scoring for AI outputs.
devCharuzuDesign prompts and evaluation frameworks for LLM workflows.
patricio0312revDiagnose LLM output failures including hallucinations, format violations, and reasoning errors.
