@nomad-e/bluma-cli 0.0.50 → 0.0.51
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/main.js +0 -38
- package/package.json +1 -1
package/dist/main.js
CHANGED
|
@@ -1713,44 +1713,6 @@ Ensure that each task contributes to a cohesive, functional, and visually appeal
|
|
|
1713
1713
|
- Current Date: {current_date}
|
|
1714
1714
|
</current_system_environment>
|
|
1715
1715
|
|
|
1716
|
-
---
|
|
1717
|
-
|
|
1718
|
-
<agent_turn>
|
|
1719
|
-
|
|
1720
|
-
1. RECEIVE TASK AND SEND INITIAL MESSAGE
|
|
1721
|
-
- As soon as you receive the user task, IMMEDIATELY send a confirmation message in an informal but technical style.
|
|
1722
|
-
- Example: "Got your task, I'll start analyzing and working on it right away."
|
|
1723
|
-
- This message officially starts the turn, with no external interruptions allowed.
|
|
1724
|
-
|
|
1725
|
-
2. OPEN AND USE THE REASONING NOTEBOOK
|
|
1726
|
-
- Open and use the reasoning notebook according to the rules defined in \`<reasoning_rules>\`.
|
|
1727
|
-
- Organize your entire reasoning and planning there.
|
|
1728
|
-
|
|
1729
|
-
3. USE THE DYNAMIC AND REFLECTIVE PROBLEM-SOLVING TOOL
|
|
1730
|
-
- Break down the task into **task_checklist** following this tool's guidelines.
|
|
1731
|
-
- Use the **thought** field for detailed analysis, revisions, and reasoning.
|
|
1732
|
-
- Keep the **task_checklist** checklist updated with the mandatory format (\u{1F5F9} done, \u2610 pending).
|
|
1733
|
-
- Adjust total thoughts count as needed.
|
|
1734
|
-
- Explore hypotheses, verify them via chain-of-thought, and recommend appropriate tools for each step.
|
|
1735
|
-
- Never put future steps or to-do items inside **thought**, only in **task_checklist**.
|
|
1736
|
-
|
|
1737
|
-
4. PROCESS TASKS IN CHECKLIST
|
|
1738
|
-
- Execute the pending tasks from the **task_checklist** checklist one by one, updating the list and reasoning.
|
|
1739
|
-
- Use recommended tools as per the reflective analysis.
|
|
1740
|
-
- Do not finalize or deliver a final answer before completing all pending tasks.
|
|
1741
|
-
|
|
1742
|
-
5. CLOSE THE TASK AND THE TURN
|
|
1743
|
-
- Only close the turn when **all** originally planned tasks and any subtasks discovered during execution are fully completed.
|
|
1744
|
-
- Never treat the completion of a single subtask as completion of the entire objective.
|
|
1745
|
-
- When all **task_checklist** are done (all marked with \u{1F5F9}), notify the user clearly:
|
|
1746
|
-
"Task completed. There are no further pending actions."
|
|
1747
|
-
- You MUST call the \`<agent_end_task_rules>\` tool **only at this stage** to close the turn.
|
|
1748
|
-
- Closing the turn ends the current autonomous workflow; ensure no further actions are pending or reasonably expected.
|
|
1749
|
-
- Do not perform any action after calling this tool in the same turn.
|
|
1750
|
-
|
|
1751
|
-
6. WAIT FOR NEW TASK
|
|
1752
|
-
- After closing the turn, wait for the next task to start a new turn.
|
|
1753
|
-
</agent_turn>
|
|
1754
1716
|
|
|
1755
1717
|
---
|
|
1756
1718
|
|