Skip to content

01: Self-Improving Debugger

Overview

An AI debugging agent that learns from each bug it encounters, building an evolving knowledge base of solutions. Each time the same error is encountered, OntoMem consolidates previous experiences using LLM-powered merging to generate better solutions.

Theme

Error Learning & Consolidation

Strategy

LLM.BALANCED merge with intelligent consolidation

Key Features

  • ✅ Multiple error encounters of the same type
  • ✅ LLM-based intelligent merging of solutions
  • ✅ Cross-encounter learning and synthesis
  • ✅ Memory persistence for future debugging sessions
  • ✅ Progressive refinement of debugging wisdom

Data Structure

DebugLog

error_id: str                      # Unique error identifier
error_type: str                    # Type of error
error_message: str                 # Error message
stack_trace: str | None            # Stack trace
solutions: list[str]               # Multiple solutions found
attempted_fixes: list[str]         # Fixes tried so far
root_cause: str | None             # Inferred root cause

Use Case

AI Development & Debugging: A debugging assistant that learns from each error, automatically consolidates solutions, and synthesizes debugging wisdom to help developers solve problems faster.

Benefits: - Automatic knowledge accumulation - Cross-project error pattern recognition - Synthesized best practices over time - Reduced time to resolution

Running the Example

cd examples/
python 01_self_improving_debugger.py

Output

Results are stored in temp/debugger_memory/: - memory.json: Consolidated error records with merged solutions - metadata.json: Schema and statistics

What You'll Learn

  1. Error Consolidation: How to merge information from multiple error encounters
  2. LLM-Powered Merging: Using LLM to intelligently synthesize conflicting solutions
  3. Cross-Encounter Learning: Building wisdom from multiple similar problems
  4. Memory Persistence: Saving and retrieving consolidated knowledge

Next Examples