Files
ragflow/rag/prompts/rank_memory.md
Kevin Hu d9fe279dde Feat: Redesign and refactor agent module (#9113)
### What problem does this PR solve?

#9082 #6365

<u> **WARNING: it's not compatible with the older version of `Agent`
module, which means that `Agent` from older versions can not work
anymore.**</u>

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00

30 lines
857 B
Markdown

**Task**: Sort the tool call results based on relevance to the overall goal and current sub-goal. Return ONLY a sorted list of indices (0-indexed).
**Rules**:
1. Analyze each result's contribution to both:
- The overall goal (primary priority)
- The current sub-goal (secondary priority)
2. Sort from MOST relevant (highest impact) to LEAST relevant
3. Output format: Strictly a Python-style list of integers. Example: [2, 0, 1]
🔹 Overall Goal: {{ goal }}
🔹 Sub-goal: {{ sub_goal }}
**Examples**:
🔹 Tool Response:
- index: 0
> Tokyo temperature is 78°F.
- index: 1
> Error: Authentication failed (expired API key).
- index: 2
> Available: 12 widgets in stock (max 5 per customer).
→ rank: [1,2,0]<|stop|>
**Your Turn**:
🔹 Tool Response:
{% for f in results %}
- index: f.i
> f.content
{% endfor %}