Why I Taught My AI Agent to Forget

Описание и применение

Создана локальная память навыков с механизмом забывания, которая хранит процедурные знания и избавляет агента от повторения старых ошибок, повышая эффективность и точность.

Задача

The AI assistant kept re‑proposing outdated procedures, wasting time because it lacked procedural memory and could not filter relevant knowledge.

Результат

Skillmem reduced redundant procedural suggestions, improved hit@5 to 0.871 and MRR to 0.622 on LongMemEval, and eliminated LLM calls during retrieval, making the system deterministic and fast.

Рабочий процесс

• 1. Identify procedural knowledge gaps in agent behavior.
• 2. Build skillmem to store skills with trigger, steps, outcome, lessons.
• 3. Integrate skillmem into Claude Code via hooks and MCP tools.
• 4. Configure decay and lifecycle rules for skill strength.
• 5. Benchmark retrieval against LongMemEval to validate performance.
• 6. Monitor and adjust decay parameters and retrieval settings.

Доказательства

skillmem stores exactly that shape. A skill looks like this: ...

Источник

hit@5 0.871, MRR 0.622 on LongMemEval

Источник

skillmem's write path is a SQLite INSERT plus Snowball stemming for the search index. No LLM in the loop.

Источник