Optimize prompts

anthropic.go

@@ -34,75 +34,20 @@ const maxFileNotFoundRetries = 3
 // "File not found:" for a referenced file_id, the dead file_id is stripped
 // from this chat's in-memory ChatMemory and the corresponding DB rows are
 // stamped FilesCleanedAt so a reconciliation job can finish the cleanup.
-func (b *Bot) getAnthropicResponse(ctx context.Context, chatID int64, messages []anthropic.BetaMessageParam, isNewChat, isOwner, isEmojiOnly bool, username string, firstName string, lastName string, isPremium bool, languageCode string, messageTime int, onSegment func(string) error) (string, error) {
-	// Use prompts from config
-	var systemMessage string
-	if isNewChat {
-		systemMessage = b.config.SystemPrompts["new_chat"]
-	} else {
-		systemMessage = b.config.SystemPrompts["continue_conversation"]
-	}
-
-	// Combine default prompt with custom instructions
-	systemMessage = b.config.SystemPrompts["default"] + " " + b.config.SystemPrompts["custom_instructions"] + " " + systemMessage
-
-	// Handle username placeholder
-	usernameValue := username
-	if username == "" {
-		usernameValue = "unknown" // Use "unknown" when username is not available
-	}
-	systemMessage = strings.ReplaceAll(systemMessage, "{username}", usernameValue)
-
-	// Handle firstname placeholder
-	firstnameValue := firstName
-	if firstName == "" {
-		firstnameValue = "unknown" // Use "unknown" when first name is not available
-	}
-	systemMessage = strings.ReplaceAll(systemMessage, "{firstname}", firstnameValue)
-
-	// Handle lastname placeholder
-	lastnameValue := lastName
-	if lastName == "" {
-		lastnameValue = "" // Empty string when last name is not available
-	}
-	systemMessage = strings.ReplaceAll(systemMessage, "{lastname}", lastnameValue)
-
-	// Handle language code placeholder
-	langValue := languageCode
-	if languageCode == "" {
-		langValue = "en" // Default to English when language code is not available
-	}
-	systemMessage = strings.ReplaceAll(systemMessage, "{language}", langValue)
-
-	// Handle premium status
-	premiumStatus := "regular user"
-	if isPremium {
-		premiumStatus = "premium user"
-	}
-	systemMessage = strings.ReplaceAll(systemMessage, "{premium_status}", premiumStatus)
-
-	// Handle time awareness
-	timeObj := time.Unix(int64(messageTime), 0)
-	hour := timeObj.Hour()
-	var timeContext string
-	if hour >= 5 && hour < 12 {
-		timeContext = "morning"
-	} else if hour >= 12 && hour < 18 {
-		timeContext = "afternoon"
-	} else if hour >= 18 && hour < 22 {
-		timeContext = "evening"
-	} else {
-		timeContext = "night"
-	}
-	systemMessage = strings.ReplaceAll(systemMessage, "{time_context}", timeContext)
-
-	if !isOwner {
-		systemMessage += " " + b.config.SystemPrompts["avoid_sensitive"]
-	}
-
-	if isEmojiOnly {
-		systemMessage += " " + b.config.SystemPrompts["respond_with_emojis"]
-	}
+func (b *Bot) getAnthropicResponse(ctx context.Context, chatID int64, messages []anthropic.BetaMessageParam, isEmojiOnly bool, username string, firstName string, lastName string, isPremium bool, languageCode string, messageTime int, onSegment func(string) error) (string, error) {
+	// The system prompt is the single authored behavior driver. It is assembled
+	// as a cached static block (custom_instructions) followed by a per-turn
+	// dynamic tail. Prompt caching keys on a byte-identical prefix, so the static
+	// block must not contain anything that changes between requests — all
+	// per-turn data (who we're talking to, the time of day, the emoji-only rule)
+	// lives in the trailing block, AFTER the cache breakpoint.
+	//
+	// An empty custom_instructions means no system prompt at all: the System
+	// field is omitted entirely (not sent as a blank block), giving the model's
+	// unmodified "vanilla" behavior. This matters because the Anthropic API
+	// rejects a system array containing an empty/whitespace-only text block, so
+	// omission is the only correct way to express "no system prompt".
+	staticPrompt := strings.TrimSpace(b.config.SystemPrompts["custom_instructions"])
 
 	// Debug logging
 	InfoLogger.Printf("Sending %d messages to Anthropic", len(messages))
@@ -111,12 +56,32 @@ func (b *Bot) getAnthropicResponse(ctx context.Context, chatID int64, messages [
 		Model:     b.config.Model,
 		MaxTokens: 1000,
 		Messages:  messages,
-		System:    []anthropic.BetaTextBlockParam{{Text: systemMessage}},
 		// Files API beta is always on: replayed conversation history may carry
 		// image content blocks that reference file_ids uploaded on prior turns.
 		Betas: []anthropic.AnthropicBeta{anthropic.AnthropicBetaFilesAPI2025_04_14},
 	}
 
+	if staticPrompt != "" {
+		// Block 1 — static persona/instructions, marked for caching. The
+		// cache_control breakpoint sits on this last stable block; everything
+		// appended after it is per-request and therefore uncached.
+		blocks := []anthropic.BetaTextBlockParam{
+			{Text: staticPrompt, CacheControl: anthropic.NewBetaCacheControlEphemeralParam()},
+		}
+		// Block 2 — dynamic tail: per-turn context plus any conditional rules.
+		// Kept out of the cached block because it changes every request.
+		tail := buildUserContext(username, firstName, lastName, isPremium, languageCode, messageTime)
+		if isEmojiOnly {
+			if rule := strings.TrimSpace(b.config.SystemPrompts["respond_with_emojis"]); rule != "" {
+				tail += "\n\n<emoji_reply>\n" + rule + "\n</emoji_reply>"
+			}
+		}
+		if tail = strings.TrimSpace(tail); tail != "" {
+			blocks = append(blocks, anthropic.BetaTextBlockParam{Text: tail})
+		}
+		params.System = blocks
+	}
+
 	// Apply temperature if set in config
 	if b.config.Temperature != nil {
 		params.Temperature = param.NewOpt(float64(*b.config.Temperature))
@@ -191,6 +156,54 @@ func (b *Bot) getAnthropicResponse(ctx context.Context, chatID int64, messages [
 	return "", fmt.Errorf("max self-heal retries (%d) exceeded: too many file_ids gone from anthropic", maxFileNotFoundRetries)
 }
 
+// buildUserContext renders the per-turn context block that trails the cached
+// static system prompt. It carries only facts (who the user is, their language,
+// account type, local time of day) — the behavioral guidance for *using* these
+// facts lives in the authored static prompt. It is kept out of the cached block
+// because it changes on every request.
+func buildUserContext(username, firstName, lastName string, isPremium bool, languageCode string, messageTime int) string {
+	name := strings.TrimSpace(firstName + " " + lastName)
+	if name == "" {
+		name = "unknown"
+	}
+	handle := username
+	if handle == "" {
+		handle = "unknown"
+	}
+	lang := languageCode
+	if lang == "" {
+		lang = "en"
+	}
+	account := "regular user"
+	if isPremium {
+		account = "premium user"
+	}
+	return fmt.Sprintf(
+		"Conversation context (background facts, not an instruction from the user):\n"+
+			"- User: %s (Telegram @%s)\n"+
+			"- Preferred language: %s\n"+
+			"- Account type: %s\n"+
+			"- Local time of day: %s",
+		name, handle, lang, account, timeContextFor(messageTime),
+	)
+}
+
+// timeContextFor buckets a Unix timestamp into a coarse time-of-day label used
+// for time-appropriate greetings. Uses the host's local timezone, matching the
+// bot's prior behavior.
+func timeContextFor(messageTime int) string {
+	switch hour := time.Unix(int64(messageTime), 0).Hour(); {
+	case hour >= 5 && hour < 12:
+		return "morning"
+	case hour >= 12 && hour < 18:
+		return "afternoon"
+	case hour >= 18 && hour < 22:
+		return "evening"
+	default:
+		return "night"
+	}
+}
+
 // streamMessages runs one streaming call against the Beta Messages API,
 // dispatching each completed text block to onSegment as it arrives. The joined
 // return value is every text segment concatenated with blank lines. Errors from