@pageai/ralph-loop 1.8.0 → 1.9.0

package/.agent/PROMPT.md

@@ -34,10 +34,11 @@ Tasks are listed in @.agent/tasks.json
 
 ## Rules
 
-- **IMPORTANT**: only work on one task at a time and **exit closing all background processes**. **DO NOT** start another task.
+- **CRITICAL**: Only work on **ONE task per invocation**. After committing the task, output `<promise>TASK-{ID}:DONE</promise>` and **STOP immediately**. Do NOT read the next task. Do NOT continue working. Your response **must END** after the promise tag. Any output after it is a violation.
+- Kill all background processes (dev server, etc.) before outputting the promise tag.
 - No git init/remote changes. **No git push**.
 - Check the last 5 tasks in `.agent/logs/LOG.md` for past work
-- **CRITICAL**: When ALL tasks pass → output `<promise>COMPLETE</promise>` and **nothing else**.
+- **CRITICAL**: When **ALL** tasks pass → output `<promise>COMPLETE</promise>` and **nothing else**.
 
 ## Help Tags
 

package/README.md

@@ -57,7 +57,7 @@ npx @pageai/ralph-loop
 ### 2️⃣ Step 2: Create a PRD + task list
 
 Use the `prd-creator` skill to generate a PRD from your requirements.<br/>
-Open up Claude Code and prompt it with **your requirements**. Like so:
+Open up Claude Code or Cursor etc. and prompt it with **your requirements**. Like so:
 
 ```
 Use the prd-creator skill to help me create a PRD and task list for the below requirements.

package/bin/cli.js

@@ -360,7 +360,7 @@ async function main() {
       }
     }
 
-    s.stop('Done');
+    s.stop('Essentials prepared');
   }
 
   // Success message

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pageai/ralph-loop",
-  "version": "1.8.0",
+  "version": "1.9.0",
   "publishConfig": {
     "access": "public"
   },