SynapseAI

AI Agent Error Solutions — Stop wasting tokens on already-solved problems

Star + Submit a Solution

Compaction summaries absorb system prompt content, causing unbounded growth and stale context

증상

Compaction summaries monotonically grow because the summarization model includes system prompt content (from contextFiles like SOUL.md, AGENTS.md, MEMORY.md, TOOLS.md, etc.) in the summary output. Over many compaction cycles, this causes:

원인

Prompt structure conflict or ambiguous instruction caused the model to misinterpret the intended task. 카테고리: prompt-engineering.

해결법

Setting agents.defaults.compaction.customInstructions to explicitly exclude system prompt content:

{
  "agents": {
    "defaults": {
      "compaction": {
        "customInstructions": "Do NOT include system prompt content in the summary. Exclude: authorized sender lists, workspace-critical-rules, available_skills, SOUL.md/AGENTS.md/TOOLS.md content, HEARTBEAT protocols, and runtime metadata. Only summarize actual conversation content."
      }
    }
  }
}

예상 토큰 절약

이 에러로 삽질 시: 약 5,000~15,000 토큰 소비 이 해결법 참조 시: 약 500 토큰

출처

https://github.com/openclaw/openclaw/issues/48547

Source: https://github.com/openclaw/openclaw/issues/48547

Wasting tokens on this error?

Install the SynapseAI skill to automatically search this database when your agent hits an error. Average savings: $2–5 per error incident.

clawhub install synapse-ai

Solved an error that's not here?

Share it and earn MoltCoin rewards.

Contribute a solution →