@letta-ai/letta-code 0.13.6 → 0.13.7

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@letta-ai/letta-code",
-  "version": "0.13.6",
+  "version": "0.13.7",
   "description": "Letta Code is a CLI tool for interacting with stateful Letta agents from the terminal.",
   "type": "module",
   "bin": {

package/skills/messaging-agents/scripts/continue-conversation.ts

@@ -22,6 +22,10 @@ import { readFileSync } from "node:fs";
 import { createRequire } from "node:module";
 import { homedir } from "node:os";
 import { join } from "node:path";
+import {
+  SYSTEM_REMINDER_CLOSE,
+  SYSTEM_REMINDER_OPEN,
+} from "../../../../constants";
 
 // Use createRequire for @letta-ai/letta-client so NODE_PATH is respected
 // (ES module imports don't respect NODE_PATH, but require does)
@@ -92,11 +96,11 @@ function buildSystemReminder(
   senderAgentName: string,
   senderAgentId: string,
 ): string {
-  return `<system-reminder>
+  return `${SYSTEM_REMINDER_OPEN}
 This message is from "${senderAgentName}" (agent ID: ${senderAgentId}), an agent currently running inside the Letta Code CLI (docs.letta.com/letta-code).
 The sender will only see the final message you generate (not tool calls or reasoning).
 If you need to share detailed information, include it in your response text.
-</system-reminder>
+${SYSTEM_REMINDER_CLOSE}
 
 `;
 }

package/skills/messaging-agents/scripts/start-conversation.ts

@@ -22,6 +22,10 @@ import { readFileSync } from "node:fs";
 import { createRequire } from "node:module";
 import { homedir } from "node:os";
 import { join } from "node:path";
+import {
+  SYSTEM_REMINDER_CLOSE,
+  SYSTEM_REMINDER_OPEN,
+} from "../../../../constants";
 
 // Use createRequire for @letta-ai/letta-client so NODE_PATH is respected
 // (ES module imports don't respect NODE_PATH, but require does)
@@ -92,11 +96,11 @@ function buildSystemReminder(
   senderAgentName: string,
   senderAgentId: string,
 ): string {
-  return `<system-reminder>
+  return `${SYSTEM_REMINDER_OPEN}
 This message is from "${senderAgentName}" (agent ID: ${senderAgentId}), an agent currently running inside the Letta Code CLI (docs.letta.com/letta-code).
 The sender will only see the final message you generate (not tool calls or reasoning).
 If you need to share detailed information, include it in your response text.
-</system-reminder>
+${SYSTEM_REMINDER_CLOSE}
 
 `;
 }

package/skills/working-in-parallel/SKILL.md

@@ -11,6 +11,16 @@ Git worktrees let you check out multiple branches into separate directories. Eac
 
 Learn more: [Git worktree documentation](https://git-scm.com/docs/git-worktree)
 
+## IMPORTANT: Check Project Setup First
+
+Before running ANY commands in a new worktree, check the project's setup instructions:
+
+1. **Read the README** - Usually has install/build commands
+2. **Check `claude.md` or `AGENT.md`** - Agent-specific guidance if present
+3. **Review your `project` memory block** - Contains learned project preferences
+
+Don't assume `npm` vs `bun` vs `pnpm` - **check the project first!**
+
 ## Quick Start
 
 ```bash
@@ -19,7 +29,10 @@ git worktree add -b fix/my-feature ../repo-my-feature main
 
 # Work in the worktree
 cd ../repo-my-feature
-bun install  # or npm install, pip install, etc.
+
+# CHECK PROJECT SETUP FIRST - then install dependencies
+# Read README.md or check project memory block for correct command
+bun install  # Example - verify this is correct for YOUR project!
 
 # Make changes, commit, push, PR
 git add <files>
@@ -48,8 +61,9 @@ git worktree remove <path>                 # Remove worktree
 
 ## Tips
 
+- **Check project setup docs before installing** - README, claude.md, project memory block
 - Name directories clearly: `../repo-feature-auth`, `../repo-bugfix-123`
-- Install dependencies in new worktrees (`npm install`, `bun install`, `pip install`, etc.)
+- Install dependencies using the project's package manager (check first!)
 - Push changes before removing worktrees
 
 ## Alternative: Repo Clones