diff --git a/lettabot.example.yaml b/lettabot.example.yaml
index 647a671..b1dbbe9 100644
--- a/lettabot.example.yaml
+++ b/lettabot.example.yaml
@@ -74,6 +74,10 @@ features:
     intervalMin: 30
     # skipRecentUserMin: 5   # Skip auto-heartbeats for N minutes after user message (0 disables)
   # memfs: true   # Enable memory filesystem (git-backed context repository). Syncs memory blocks to local files.
+  # display:
+  #   showToolCalls: false       # Show tool invocations in chat (e.g. "Using tool: Read (file_path: ...)")
+  #   showReasoning: false       # Show agent reasoning/thinking in chat
+  #   reasoningMaxChars: 0        # Truncate reasoning to N chars (0 = no limit, default)
 
 # Attachment handling (defaults to 20MB if omitted)
 # attachments:
diff --git a/src/config/types.ts b/src/config/types.ts
index 9e73b7e..06a2210 100644
--- a/src/config/types.ts
+++ b/src/config/types.ts
@@ -25,6 +25,18 @@ export function serverModeLabel(mode?: ServerMode): string {
   return canonicalizeServerMode(mode);
 }
 
+/**
+ * Display configuration for tool calls and reasoning in channel output.
+ */
+export interface DisplayConfig {
+  /** Show tool invocations in channel output (default: false) */
+  showToolCalls?: boolean;
+  /** Show agent reasoning/thinking in channel output (default: false) */
+  showReasoning?: boolean;
+  /** Truncate reasoning to N characters (default: 0 = no limit) */
+  reasoningMaxChars?: number;
+}
+
 /**
  * Configuration for a single agent in multi-agent mode.
  * Each agent has its own name, channels, and features.
@@ -65,6 +77,7 @@ export interface AgentConfig {
     };
     memfs?: boolean;          // Enable memory filesystem (git-backed context repository) for SDK sessions
     maxToolCalls?: number;
+    display?: DisplayConfig;
   };
   /** Polling config */
   polling?: PollingYamlConfig;
@@ -138,6 +151,7 @@ export interface LettaBotConfig {
     inlineImages?: boolean;   // Send images directly to the LLM (default: true). Set false to only send file paths.
     memfs?: boolean;          // Enable memory filesystem (git-backed context repository) for SDK sessions
     maxToolCalls?: number;  // Abort if agent calls this many tools in one turn (default: 100)
+    display?: DisplayConfig;  // Show tool calls / reasoning in channel output
   };
 
   // Polling - system-level background checks (Gmail, etc.)
diff --git a/src/core/bot.ts b/src/core/bot.ts
index 1c4b096..bf18464 100644
--- a/src/core/bot.ts
+++ b/src/core/bot.ts
@@ -211,6 +211,50 @@ export class LettaBot implements AgentSession {
     return `${this.config.displayName}: ${text}`;
   }
 
+  /**
+   * Format a tool call for channel display.
+   * Shows tool name + abbreviated key parameters.
+   */
+  private formatToolCallDisplay(streamMsg: StreamMsg): string {
+    const name = streamMsg.toolName || 'unknown';
+    const params = this.abbreviateToolInput(streamMsg);
+    return params ? `> **Tool:** ${name} (${params})` : `> **Tool:** ${name}`;
+  }
+
+  /**
+   * Extract a brief parameter summary from a tool call's input.
+   */
+  private abbreviateToolInput(streamMsg: StreamMsg): string {
+    const input = streamMsg.toolInput as Record<string, unknown> | undefined;
+    if (!input || typeof input !== 'object') return '';
+    const entries = Object.entries(input).slice(0, 2);
+    return entries
+      .map(([k, v]) => {
+        let str: string;
+        try {
+          str = typeof v === 'string' ? v : (JSON.stringify(v) ?? String(v));
+        } catch {
+          str = String(v);
+        }
+        const truncated = str.length > 80 ? str.slice(0, 77) + '...' : str;
+        return `${k}: ${truncated}`;
+      })
+      .join(', ');
+  }
+
+  /**
+   * Format reasoning text for channel display, respecting truncation config.
+   */
+  private formatReasoningDisplay(text: string): string {
+    const maxChars = this.config.display?.reasoningMaxChars ?? 0;
+    const truncated = maxChars > 0 && text.length > maxChars
+      ? text.slice(0, maxChars) + '...'
+      : text;
+    // Prefix every line with "> " so the whole block renders as a blockquote
+    const lines = truncated.split('\n').map(line => `> ${line}`);
+    return `> **Thinking**\n${lines.join('\n')}`;
+  }
+
   // =========================================================================
   // Session options (shared by processMessage and sendToAgent)
   // =========================================================================
@@ -1064,6 +1108,7 @@ export class LettaBot implements AgentSession {
       let sawNonAssistantSinceLastUuid = false;
       let lastErrorDetail: { message: string; stopReason: string; apiError?: Record<string, unknown> } | null = null;
       let retryInfo: { attempt: number; maxAttempts: number; reason: string } | null = null;
+      let reasoningBuffer = '';
       const msgTypeCounts: Record<string, number> = {};
       
       const finalizeMessage = async () => {
@@ -1122,6 +1167,20 @@ export class LettaBot implements AgentSession {
           if (lastMsgType && lastMsgType !== streamMsg.type && response.trim() && streamMsg.type !== 'result') {
             await finalizeMessage();
           }
+
+          // Flush reasoning buffer when type changes away from reasoning
+          if (lastMsgType === 'reasoning' && streamMsg.type !== 'reasoning' && reasoningBuffer.trim()) {
+            if (this.config.display?.showReasoning && !suppressDelivery) {
+              try {
+                const text = this.formatReasoningDisplay(reasoningBuffer);
+                await adapter.sendMessage({ chatId: msg.chatId, text, threadId: msg.threadId });
+                sentAnyMessage = true;
+              } catch (err) {
+                console.warn('[Bot] Failed to send reasoning display:', err instanceof Error ? err.message : err);
+              }
+            }
+            reasoningBuffer = '';
+          }
           
           // Tool loop detection
           const maxToolCalls = this.config.maxToolCalls ?? 100;
@@ -1137,14 +1196,30 @@ export class LettaBot implements AgentSession {
             this.syncTodoToolCall(streamMsg);
             console.log(`[Stream] >>> TOOL CALL: ${streamMsg.toolName || 'unknown'} (id: ${streamMsg.toolCallId?.slice(0, 12) || '?'})`);
             sawNonAssistantSinceLastUuid = true;
+            // Display tool call in channel if configured
+            if (this.config.display?.showToolCalls && !suppressDelivery) {
+              try {
+                const text = this.formatToolCallDisplay(streamMsg);
+                await adapter.sendMessage({ chatId: msg.chatId, text, threadId: msg.threadId });
+                sentAnyMessage = true;
+              } catch (err) {
+                console.warn('[Bot] Failed to send tool call display:', err instanceof Error ? err.message : err);
+              }
+            }
           } else if (streamMsg.type === 'tool_result') {
             console.log(`[Stream] <<< TOOL RESULT: error=${streamMsg.isError}, len=${(streamMsg as any).content?.length || 0}`);
             sawNonAssistantSinceLastUuid = true;
           } else if (streamMsg.type === 'assistant' && lastMsgType !== 'assistant') {
             console.log(`[Bot] Generating response...`);
-          } else if (streamMsg.type === 'reasoning' && lastMsgType !== 'reasoning') {
-            console.log(`[Bot] Reasoning...`);
+          } else if (streamMsg.type === 'reasoning') {
+            if (lastMsgType !== 'reasoning') {
+              console.log(`[Bot] Reasoning...`);
+            }
             sawNonAssistantSinceLastUuid = true;
+            // Accumulate reasoning content for display
+            if (this.config.display?.showReasoning) {
+              reasoningBuffer += streamMsg.content || '';
+            }
           } else if (streamMsg.type === 'error') {
             // SDK now surfaces error detail that was previously dropped.
             // Store for use in the user-facing error message.
diff --git a/src/core/types.ts b/src/core/types.ts
index 4f35ad7..0d18eb0 100644
--- a/src/core/types.ts
+++ b/src/core/types.ts
@@ -128,6 +128,11 @@ export interface BotConfig {
 
   // Display
   displayName?: string; // Prefix outbound messages (e.g. "💜 Signo")
+  display?: {
+    showToolCalls?: boolean;      // Show tool invocations in channel output
+    showReasoning?: boolean;      // Show agent reasoning/thinking in channel output
+    reasoningMaxChars?: number;   // Truncate reasoning to N chars (default: 0 = no limit)
+  };
 
   // Skills
   skills?: SkillsConfig;
diff --git a/src/main.ts b/src/main.ts
index 7230618..936c3a1 100644
--- a/src/main.ts
+++ b/src/main.ts
@@ -531,6 +531,7 @@ async function main() {
       displayName: agentConfig.displayName,
       maxToolCalls: agentConfig.features?.maxToolCalls,
       memfs: resolvedMemfs,
+      display: agentConfig.features?.display,
       conversationMode: agentConfig.conversations?.mode || 'shared',
       heartbeatConversation: agentConfig.conversations?.heartbeat || 'last-active',
       skills: {