@townco/agent 0.1.46 → 0.1.48

package/dist/acp-server/adapter.js

@@ -17,15 +17,25 @@ function createContextSnapshot(messageCount, timestamp, previousContext) {
     if (previousContext) {
         // Start with all messages from previous context
         messages.push(...previousContext.messages);
-        // Find the highest pointer index in previous context
-        let maxPointerIndex = -1;
-        for (const entry of previousContext.messages) {
-            if (entry.type === "pointer" && entry.index > maxPointerIndex) {
-                maxPointerIndex = entry.index;
+        // Determine the starting index for new pointers
+        let startPointerIndex;
+        if (previousContext.compactedUpTo !== undefined) {
+            // If previous context was a compaction, start adding pointers
+            // for messages after the compaction boundary
+            startPointerIndex = previousContext.compactedUpTo + 1;
+        }
+        else {
+            // Find the highest pointer index in previous context
+            let maxPointerIndex = -1;
+            for (const entry of previousContext.messages) {
+                if (entry.type === "pointer" && entry.index > maxPointerIndex) {
+                    maxPointerIndex = entry.index;
+                }
             }
+            startPointerIndex = maxPointerIndex + 1;
         }
         // Add pointers for any new messages since the previous context
-        for (let i = maxPointerIndex + 1; i < messageCount; i++) {
+        for (let i = startPointerIndex; i < messageCount; i++) {
             messages.push({ type: "pointer", index: i });
         }
     }
@@ -294,6 +304,18 @@ export class AgentAcpAdapter {
                 if (hookResult.newContextEntries.length > 0) {
                     logger.info(`Appending ${hookResult.newContextEntries.length} new context entries from hooks`);
                     session.context.push(...hookResult.newContextEntries);
+                    // Save session immediately after hooks to persist compacted context
+                    if (this.storage) {
+                        try {
+                            await this.storage.saveSession(params.sessionId, session.messages, session.context);
+                            logger.info("Session saved after hook execution with new context entries");
+                        }
+                        catch (error) {
+                            logger.error(`Failed to save session ${params.sessionId} after hook execution`, {
+                                error: error instanceof Error ? error.message : String(error),
+                            });
+                        }
+                    }
                 }
             }
             // Resolve context to messages for agent invocation

package/dist/acp-server/session-storage.d.ts

@@ -44,6 +44,12 @@ export type ContextMessageEntry = MessagePointer | FullMessageEntry;
 export interface ContextEntry {
     timestamp: string;
     messages: ContextMessageEntry[];
+    /**
+     * When set, indicates this context entry represents a compaction
+     * and all messages up to and including this index have been
+     * compacted into the full message(s) in this entry
+     */
+    compactedUpTo?: number | undefined;
 }
 /**
  * Session metadata

package/dist/acp-server/session-storage.js

@@ -55,6 +55,7 @@ const contextMessageEntrySchema = z.discriminatedUnion("type", [
 const contextEntrySchema = z.object({
     timestamp: z.string(),
     messages: z.array(contextMessageEntrySchema),
+    compactedUpTo: z.number().optional(),
 });
 const sessionMetadataSchema = z.object({
     createdAt: z.string(),

package/dist/runner/hooks/predefined/compaction-tool.d.ts

@@ -1,6 +1,6 @@
 import { type HookCallback } from "../types";
 /**
- * Placeholder compaction tool for testing hooks
- * Logs the context and returns null (no new context entry)
+ * Compaction tool that uses an LLM to summarize conversation history
+ * when context size reaches the configured threshold
  */
 export declare const compactionTool: HookCallback;

package/dist/runner/hooks/predefined/compaction-tool.js

@@ -1,9 +1,11 @@
+import { ChatAnthropic } from "@langchain/anthropic";
+import { HumanMessage, SystemMessage } from "@langchain/core/messages";
 import { createLogger } from "@townco/core";
 import { createContextEntry, createFullMessageEntry, } from "../types";
 const logger = createLogger("compaction-tool");
 /**
- * Placeholder compaction tool for testing hooks
- * Logs the context and returns null (no new context entry)
+ * Compaction tool that uses an LLM to summarize conversation history
+ * when context size reaches the configured threshold
  */
 export const compactionTool = async (ctx) => {
     logger.info("Compaction tool triggered", {
@@ -14,29 +16,103 @@ export const compactionTool = async (ctx) => {
         totalMessages: ctx.session.messages.length,
         model: ctx.model,
     });
-    // TODO: Implement actual compaction logic
-    // Example: Create a new context entry with a summary message
-    // const summaryEntry = createFullMessageEntry(
-    //   "assistant",
-    //   "Summary of previous conversation: ..."
-    // );
-    //
-    // const newContextEntry = createContextEntry([summaryEntry]);
-    //
-    // return {
-    //   newContextEntry,
-    //   metadata: {
-    //     action: "compacted",
-    //     messagesRemoved: ctx.session.messages.length - 1,
-    //     tokensSaved: 1000,
-    //   },
-    // };
-    // For now, just return null (no new context entry)
-    return {
-        newContextEntry: null,
-        metadata: {
-            action: "placeholder",
-            message: "Compaction tool called but no action taken (placeholder)",
-        },
-    };
+    try {
+        // Create the LLM client using the same model as the agent
+        const model = new ChatAnthropic({
+            model: ctx.model,
+            temperature: 0,
+        });
+        // Build the conversation history to compact
+        const messagesToCompact = ctx.session.messages;
+        // Convert session messages to text for context
+        const conversationText = messagesToCompact
+            .map((msg) => {
+            const textContent = msg.content
+                .filter((block) => block.type === "text")
+                .map((block) => block.text)
+                .join("\n");
+            return `${msg.role.toUpperCase()}:\n${textContent}`;
+        })
+            .join("\n\n---\n\n");
+        // Create system prompt for compaction
+        const systemPrompt = new SystemMessage("You are a helpful AI assistant tasked with summarizing conversations.");
+        // Create detailed compaction instructions with a generic, domain-agnostic approach
+        const userPrompt = `Your task is to create a detailed summary of the conversation so far, paying close attention to the user's explicit requests and your previous actions.
+This summary should be thorough in capturing important details, decisions, and context that would be essential for continuing the conversation/task without losing important information.
+
+Before providing your final summary, wrap your analysis in <analysis> tags to organize your thoughts and ensure you've covered all necessary points. In your analysis process:
+
+1. Chronologically analyze each message and section of the conversation. For each section thoroughly identify:
+   - The user's explicit requests and intents
+   - Your approach to addressing the user's requests
+   - Key decisions and important concepts discussed
+   - Specific details that are important to remember
+   - Challenges encountered and how they were addressed
+   - Pay special attention to specific user feedback, especially if the user told you to do something differently
+2. Double-check for accuracy and completeness, addressing each required element thoroughly
+
+Your summary should include the following sections:
+
+1. Primary Request and Intent: Capture all of the user's explicit requests and intents in detail
+2. Key Topics and Concepts: List all important topics, concepts, and themes discussed in the conversation
+3. Important Details and Information: Document specific details, information, or content that was shared, examined, or created. Pay special attention to the most recent messages and include important details where applicable
+4. Challenges and Solutions: List any challenges, issues, or obstacles that came up and how they were addressed. Pay special attention to specific user feedback, especially if the user asked for a different approach
+5. Problem Solving: Document problems solved and any ongoing work or troubleshooting efforts
+6. All User Messages: List ALL user messages (excluding tool results). These are critical for understanding the user's feedback and changing intent
+7. Pending Tasks: Outline any pending tasks that you have explicitly been asked to work on
+8. Current Work: Describe in detail precisely what was being worked on immediately before this summary, paying special attention to the most recent messages from both user and assistant
+9. Optional Next Step: List the next step related to the most recent work. IMPORTANT: ensure this step is DIRECTLY in line with the user's most recent explicit requests and the task you were working on. If the last task was concluded, only list next steps if they are explicitly in line with the user's request. Do not start on tangential requests or old requests that were already completed without confirming with the user first.
+   If there is a next step, include direct quotes from the most recent conversation showing exactly what task you were working on and where you left off. This should be verbatim to ensure there's no drift in task interpretation.
+
+Here's the conversation to summarize:
+
+${conversationText}
+
+Please provide your summary based on the conversation above, following this structure and ensuring precision and thoroughness in your response.`;
+        const userMessage = new HumanMessage(userPrompt);
+        // Invoke the LLM
+        logger.info("Invoking LLM for compaction summary");
+        const response = await model.invoke([systemPrompt, userMessage]);
+        // Extract the summary text from the response
+        const summaryText = typeof response.content === "string"
+            ? response.content
+            : Array.isArray(response.content)
+                ? response.content
+                    .filter((block) => typeof block === "object" && block !== null && "text" in block)
+                    .map((block) => block.text)
+                    .join("\n")
+                : "Failed to extract summary";
+        logger.info("Generated compaction summary", {
+            originalMessages: messagesToCompact.length,
+            summaryLength: summaryText.length,
+            estimatedTokensSaved: Math.round(ctx.currentTokens * 0.7),
+        });
+        // Create a new context entry with the summary
+        const summaryEntry = createFullMessageEntry("user", `This session is being continued from a previous conversation that ran out of context. The conversation is summarized below:\n${summaryText}`);
+        // Set compactedUpTo to indicate all messages have been compacted into the summary
+        const lastMessageIndex = messagesToCompact.length - 1;
+        const newContextEntry = createContextEntry([summaryEntry], undefined, lastMessageIndex);
+        return {
+            newContextEntry,
+            metadata: {
+                action: "compacted",
+                messagesRemoved: messagesToCompact.length - 1,
+                tokensSaved: Math.round(ctx.currentTokens * 0.7),
+                summaryGenerated: true,
+            },
+        };
+    }
+    catch (error) {
+        logger.error("Compaction tool error", {
+            error: error instanceof Error ? error.message : String(error),
+            stack: error instanceof Error ? error.stack : undefined,
+        });
+        return {
+            newContextEntry: null,
+            metadata: {
+                action: "failed",
+                error: error instanceof Error ? error.message : String(error),
+            },
+        };
+    }
 };

package/dist/runner/hooks/types.d.ts

@@ -106,7 +106,7 @@ export declare function createContextEntry(messages: Array<{
 } | {
     type: "full";
     message: SessionMessage;
-}>, timestamp?: string): ContextEntry;
+}>, timestamp?: string, compactedUpTo?: number): ContextEntry;
 /**
  * Helper function to create a full message entry for context
  * Use this when hooks need to inject new messages into context

package/dist/runner/hooks/types.js

@@ -2,11 +2,15 @@
  * Helper function to create a new context entry
  * Use this when hooks want to create a new context snapshot
  */
-export function createContextEntry(messages, timestamp) {
-    return {
+export function createContextEntry(messages, timestamp, compactedUpTo) {
+    const entry = {
         timestamp: timestamp || new Date().toISOString(),
         messages,
     };
+    if (compactedUpTo !== undefined) {
+        entry.compactedUpTo = compactedUpTo;
+    }
+    return entry;
 }
 /**
  * Helper function to create a full message entry for context