compaction

type PruneConfig struct {
	// MaxToolResultBytes is per-result truncation limit.
	MaxToolResultBytes int `json:"maxToolResultBytes"`
}

type SummaryConfig struct {
	// MinMessages is an optional pre-gate when Force is false. Zero disables it.
	MinMessages int `json:"minMessages"`
	// KeepRecentTokens is the recent context budget kept verbatim (Pi-compatible).
	KeepRecentTokens int `json:"keepRecentTokens"`
	// ReserveTokens caps summarization output size.
	ReserveTokens int `json:"reserveTokens"`
	// KeepRecent is deprecated; use keepRecentTokens.
	KeepRecent int `json:"keepRecent"`
	// SummaryModel is model used for the summary; defaults to the agent's model.
	SummaryModel string `json:"summaryModel"`
	// SummaryPrompt overrides the built-in summarisation instruction.
	SummaryPrompt string                     `json:"summaryPrompt"`
	Retry         *capcompaction.RetryConfig `json:"retry,omitempty"`
}

type SummaryDeps struct {
	LLM agentkit.LLMProvider `json:"llm"`
}

type TokenLimitConfig struct {
	// MaxTokens is absolute trigger; takes precedence over ContextWindow.
	MaxTokens int `json:"maxTokens"`
	// ContextWindow is model context size; the trigger becomes ContextWindow × TriggerRatio.
	ContextWindow int `json:"contextWindow"`
	// TriggerRatio is fraction of ContextWindow that trips compaction, default 0.7 — leaving room for the reply plus the next tool result.
	TriggerRatio float64 `json:"triggerRatio"`
	// CharsPerToken calibrates the fallback estimate used before the provider reports usage; default 4 (English prose), lower it for CJK.
	CharsPerToken int `json:"charsPerToken"`
}

type TokenLimitDeps struct {
	// Services run only once the threshold is crossed, with Force set.
	Services []compaction.Service `json:"services"`
}