|
30 | 30 | 5. Summarize all child memory items into one memory item. |
31 | 31 |
|
32 | 32 | Language rules: |
33 | | -- The `key`, `value`, `tags`, `summary` fields must match the mostly used language of the input memory items. |
| 33 | +- The `key`, `value`, `tags`, `summary` fields must match the mostly used language of the input memory items. **如果输入是中文,请输出中文** |
34 | 34 | - Keep `memory_type` in English. |
35 | 35 |
|
36 | 36 | Return valid JSON: |
37 | 37 | { |
38 | 38 | "key": <string, a unique, concise memory title>, |
39 | 39 | "memory_type": <string, Either "LongTermMemory" or "UserMemory">, |
40 | | - "value": <A detailed, self-contained, and unambiguous memory statement — written in English if the input conversation is in English, or in Chinese if the conversation is in Chinese>, |
| 40 | + "value": <A detailed, self-contained, and unambiguous memory statement — written in English if the input memory items are in English, or in Chinese if the input is in Chinese>, |
41 | 41 | "tags": <A list of relevant thematic keywords (e.g., ["deadline", "team", "planning"])>, |
42 | 42 | "summary": <a natural paragraph summarizing the above memories from user's perspective, 120–200 words, same language as the input> |
43 | 43 | } |
|
59 | 59 | Example: If you have items about a project across multiple phases, group them by milestone, team, or event. |
60 | 60 |
|
61 | 61 | Language rules: |
62 | | -- The `theme` fields must match the mostly used language of the clustered memories. |
| 62 | +- The `theme` fields must match the mostly used language of the clustered memories. **如果输入是中文,请输出中文** |
63 | 63 |
|
64 | 64 | Return valid JSON: |
65 | 65 | { |
|
0 commit comments