Skip to content

Commit 1379f39

Browse files
authored
use default case (#6)
* sepcs * llm-based ontology generate * generate ontology and run baseline * specs * market tech spaces * market entry strategy * ontology auto-fix and warning * specs * reality observation * add reality path * curve edge * state transition * base bend * clear submodule * specs * specs * shenda case * founder ontology generated * founder graph * founder graph * founder graph * refactor prompt * strategic actor * graph layout * fix gramm error * rollback spring layout * fix edge label * founder ontology * founder view * JSON gen and pre * formation and insight * formation and insight * founder insight * UI layout * upgrade depends * Added Dev Container Folder * rollback python version * streamlit config * product view * founder app * analysis status * Founder APP * remove embedding * i18n * LLM language * unknown * timeline descript * outcome output * timeline * load env * open source plan * open source * open source * prompt tiering * prompt refactor * CLI validated * restore case analysis * fix events nodes * founder validated * founder output folder * founder output folder * remove solution submodule * move solution to specs * strategic actor * specs * unit test passed * CLI validated * validated and refactor * remove founder insight * remove founder insight * remove founder * specs * remove founder * remove enhancement * specs * refactor actor ontologyy * refactor actor * restor actor relationship * actor type * remeve public suffix * fix background_facts * added strategic style * added strategic style * show strategic style * actor profile * fix strategic actor profile * actor persona * remove constraints * remove embedding * refactor ingest module * strategic actor docs * strategic actor docs * 007 founder opensource (#15) * sepcs * llm-based ontology generate * generate ontology and run baseline * specs * market tech spaces * market entry strategy * ontology auto-fix and warning * specs * reality observation * add reality path * curve edge * state transition * base bend * clear submodule * specs * specs * shenda case * founder ontology generated * founder graph * founder graph * founder graph * refactor prompt * strategic actor * graph layout * fix gramm error * rollback spring layout * fix edge label * founder ontology * founder view * JSON gen and pre * formation and insight * formation and insight * founder insight * UI layout * upgrade depends * rollback python version * streamlit config * product view * founder app * analysis status * Founder APP * remove embedding * i18n * LLM language * unknown * timeline descript * outcome output * timeline * load env * open source plan * open source * open source * prompt tiering * prompt refactor * CLI validated * restore case analysis * fix events nodes * founder validated * founder output folder * founder output folder * remove solution submodule * move solution to specs * strategic actor * specs * unit test passed * CLI validated * validated and refactor * remove founder insight * remove founder insight * remove founder * specs * remove founder * remove enhancement * specs * refactor actor ontologyy * refactor actor * restor actor relationship * actor type * remeve public suffix * fix background_facts * added strategic style * added strategic style * show strategic style * actor profile * fix strategic actor profile * actor persona * remove constraints * remove embedding * refactor ingest module * strategic actor docs * strategic actor docs * config and llm * use default case * remove container * use default case
1 parent 06324ed commit 1379f39

1 file changed

Lines changed: 24 additions & 14 deletions

File tree

docs/quick-start-llm.md

Lines changed: 24 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11

22
# 快速开始:使用 LLM 生成
33

4-
Omen `0.2.0` 版本起开始集成大语言模型增强生成。本文档说明如何配置大模型,通过 `Strategic Actor` CLI 示例生成产出物、完成结果检查。
4+
Omen `0.2.0` 版本起集成大语言模型增强生成。本文档说明如何配置大模型,通过 `Strategic Actor` CLI 示例生成产出物、完成结果检查。
55

66
## 1. 准备运行
77

@@ -37,49 +37,59 @@ cp config/llm.example.toml config/llm.toml
3737

3838
编辑 `config/llm.toml`,确认 `provider``model` 以及引用的环境变量名是否正确。
3939

40-
## 2. 准备输入文档
40+
## 2. 运行示例
4141

42-
将案例文档放在 `cases/actors/`,例如:`cases/actors/x-developer.md`
42+
示例文档在 `cases/actors/`,例如:`cases/actors/chen-jiaxing.md`
4343

44-
命令中的 `--doc x-developer` 会解析到上述文件。
45-
46-
### 运行 CLI 生成
44+
运行 CLI 命令,生成产出物:
4745

4846
```bash
49-
omen analyze actor --doc x-developer --config config/llm.toml
47+
omen analyze actor --doc chen-jiaxing
5048
```
5149

52-
默认输出目录:
53-
54-
- `output/actors/x-developer/`
50+
默认输出目录:`output/actors/chen-jiaxing/`
5551

5652
## 3. 检查生成结果
5753

5854
先检查目录与文件是否存在:
5955

6056
```bash
61-
ls -lah output/actors/x-developer/
57+
ls -lah output/actors/chen-jiaxing/
6258
```
6359

64-
期望至少包含
60+
应包含以下文件
6561

6662
- `strategy_ontology.json`
6763
- `actor_ontology.json`
6864
- `analyze_status.json`
6965
- `analyze_persona.json`
7066
- `generation.json`
7167

72-
再执行结构校验
68+
执行结构校验
7369

7470
```bash
75-
omen validate actor --doc x-developer --output-dir output/actors
71+
omen validate actor --doc chen-jiaxing --output-dir output/actors
7672
```
7773

7874
判定标准:
7975

8076
- 输出 `status=pass`:可进入后续 UI 展示或下游流程
8177
- 输出 `status=fail`:根据 `errors` 字段逐项修复输入文档或配置后重试
8278

79+
## 4. 可视化展示
80+
81+
生成的 JSON 文件可通过 Omen UI 展示:
82+
83+
```bash
84+
streamlit run app/strategic_actor.py
85+
```
86+
87+
访问 `http://localhost:8501`,选择对应的输出目录,即可查看生成的战略本体、角色本体等信息。
88+
89+
![](assets/images/streamlit-strategic-actor-persona.png)
90+
91+
![](assets/images/streamlit-strategic-actor-graph-timeline.png)
92+
8393
---
8494

8595
## 常见问题

0 commit comments

Comments
 (0)