Unverified50% confidenceFactExact time
Tool outputs are repeatedly read and repeatedly billed in every subsequent round of conversation until the context is compressed.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDuring multi-turn tool calls, each tool execution result is appended to the conversation history as a new tool_result message, and the model reads the outputs of previous steps from the conversation context rather than retaining intermediate values in internal memory.79% similarUnverifiedThe /compact command generates a summary of the current conversation and adds it to a brand-new conversation; it is recommended to manually compress at 70%–90% usage.75% similarUnverifiedThe /compact command works by summarizing and compressing the current session, retaining key decisions and code change records while discarding redundant intermediate discussions.74% similarUnverifiedCodeCook reduces token costs by summarizing and compressing earlier conversation turns and large-volume tool outputs.73% similarUnverifiedSaved Skills appear in the conversation list and can be invoked at any time, enabling a build-once, reuse-many-times workflow.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1004242API
curl https://kongchang.com/api/v1/knowledge/claims/1004242MCP
get_claim(id=1004242)