github/ChatDev
1
0
Fork 0
mirror of https://github.com/OpenBMB/ChatDev.git synced 2026-04-25 11:18:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix typos

Browse Source
This commit is contained in:
NA-Wen 2026-03-17 10:14:32 +08:00
parent 033617e4ba
commit c2dd7abd55
5 changed files with 334 additions and 407 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README-zh.md
View File

@ -180,7 +180,7 @@ OpenClaw 可以与 ChatDev 集成，通过 **调用已有的 agent 团队**，
2. 为你的 OpenClaw 实例安装所需的技能：
```bash
clawdbot install chatdev
clawdhub install chatdev
```
3. 让 OpenClaw 创建一个 ChatDev 工作流。例如：

2
README.md
View File

@ -188,7 +188,7 @@ To get started:
1. Start the ChatDev 2.0 backend.
2. Install the required skills for your OpenClaw instance:
```bash
clawdbot install chatdev
clawdhub install chatdev
```
3. Ask your OpenClaw to create a ChatDev workflow. For example:

405
skills.md
View File

@ -1,405 +0,0 @@
---
name: chatdev
description: Invoke ChatDev ability units (workflows) via local API (port 6400). Use when the user needs specialized agent workflows like paper reviews (arXiv), data visualization (from CSV), or when they mention "ChatDev", "ability units", or "workflows" at http://127.0.0.1:6400.
---
# ChatDev Ability Units
Ability units represent capability units. You can browse what's available, inspect
required args, run them, or upload new ones. Use these endpoints when you need to
invoke a predefined ability unit (e.g., paper review, data visualization) via the
local API.
Run ability units via the local API and return `final_message` from the JSON response.
## API endpoints
1) Browse ability units
- Method: GET
- URL: http://127.0.0.1:6400/api/workflows
- Example:
```
curl --noproxy 127.0.0.1 -v http://127.0.0.1:6400/api/workflows
```
2) Get ability unit raw content
- Method: GET
- URL: http://127.0.0.1:6400/api/workflows/<filename>/get
- Purpose: fetch the raw YAML content for an ability unit file.
- Example:
```
curl --noproxy 127.0.0.1 -v \
http://127.0.0.1:6400/api/workflows/test.yaml/get
```
3) Get ability unit args
- Method: GET
- URL: http://127.0.0.1:6400/api/workflows/<ability_unit>.yaml/args
- Purpose: fetch input parameter schema for an ability unit.
- Example:
```
curl --noproxy 127.0.0.1 -v -X GET \
http://127.0.0.1:6400/api/workflows/test.yaml/args
```
4) Run ability unit
- Method: POST
- URL: http://127.0.0.1:6400/api/workflow/run
- Content-Type: application/json
- SSE support: set `Accept: text/event-stream` to stream events (`started`, `log`,
`completed`, `error`). Omit the header to get the normal JSON response.
- Output: parse JSON and return `final_message`. If missing, report failure and
include any error fields. After you get `output_dir`, move it to your working
directory.
- Available ability units:
1) Paper Review
- yaml_file: `yaml_instance/paper_review.yaml`
- task_prompt: arXiv URL/ID or paper title.
- Purpose: fetch content and deliver technical + writing reviews.
- Example:
```
curl --noproxy 127.0.0.1 -v -X POST http://127.0.0.1:6400/api/workflow/run \
-H "Content-Type: application/json" \
-H "Accept: text/event-stream" \
-d '{
"yaml_file": "yaml_instance/paper_review.yaml",
"task_prompt": "https://arxiv.org/abs/1706.03762"
}'
```
2) Data Visualization
- yaml_file: `yaml_instance/data_visualization_basic.yaml`
- task_prompt: dataset description + analysis goals.
- attachments: absolute paths to CSV files.
- Purpose: profile, clean, plan 4-6 charts, iterate visualizations.
- Example:
```
curl --noproxy 127.0.0.1 -v -X POST http://127.0.0.1:6400/api/workflow/run \
-H "Content-Type: application/json" \
-d '{
"yaml_file": "yaml_instance/data_visualization_basic.yaml",
"task_prompt": "please analyze and visualize the data",
"attachments": ["/Users/yufan/Projects/bugfix/ChatDev/sample_sales.csv"]
}'
```
5) Upload ability unit
- Method: POST
- URL: http://127.0.0.1:6400/api/workflows/upload/content
- Content-Type: application/json
- Format notes: `filename` should end with `.yaml`; `content` must be a valid YAML
string that includes `version`, `vars`, and a `graph` with `id`, `start`,
`nodes`, and `edges` (as in the example).
- Tip: if uploads keep failing with format errors, call the "get ability unit raw content" endpoint to inspect
an existing ability unit's structure and copy its format.
- Example:
```
curl --noproxy 127.0.0.1 -v -X POST \
http://127.0.0.1:6400/api/workflows/upload/content \
-H "Content-Type: application/json" \
-d @- <<'EOF'
{
"filename": "test.yaml",
"content": "version: 0.4.0\nvars: {}\ngraph:\n id: paper_review\n description: Three agents collaboratively review academic papers from different perspectives. Input can be an arxiv URL or paper title.\n is_majority_voting: false\n start:\n - Paper Fetcher\n nodes:\n - id: Paper Fetcher\n type: agent\n config:\n name: gpt-4o\n provider: openai\n role: |\n You are a paper content fetcher.\n base_url: ${BASE_URL}\n api_key: ${API_KEY}\n description: Fetches paper content from arxiv\n context_window: 0\n edges: []"
}
EOF
```
6) Update ability unit
- Method: PUT
- URL: http://127.0.0.1:6400/api/workflows/<filename>/update
- Content-Type: application/json
- Format notes: same payload as upload (`filename` + YAML `content` string).
- Example:
```
curl --noproxy 127.0.0.1 -v -X PUT \
http://127.0.0.1:6400/api/workflows/test.yaml/update \
-H "Content-Type: application/json" \
-d @- <<'EOF'
{
"filename": "test.yaml",
"content": "version: 0.4.0\nvars: {}\ngraph:\n id: paper_review\n description: Update example.\n is_majority_voting: false\n start:\n - Paper Fetcher\n nodes:\n - id: Paper Fetcher\n type: agent\n config:\n name: gpt-4o\n provider: openai\n role: |\n You are a paper content fetcher.\n base_url: ${BASE_URL}\n api_key: ${API_KEY}\n description: Fetches paper content from arxiv\n context_window: 0\n edges: []"
}
EOF
```
7) Rename ability unit
- Method: POST
- URL: http://127.0.0.1:6400/api/workflows/<filename>/rename
- Content-Type: application/json
- Body: `{ "new_filename": "new_name.yaml" }`
- Example:
```
curl --noproxy 127.0.0.1 -v -X POST \
http://127.0.0.1:6400/api/workflows/test.yaml/rename \
-H "Content-Type: application/json" \
-d '{"new_filename":"renamed.yaml"}'
```
8) Copy ability unit
- Method: POST
- URL: http://127.0.0.1:6400/api/workflows/<filename>/copy
- Content-Type: application/json
- Body: `{ "new_filename": "copy.yaml" }`
- Example:
```
curl --noproxy 127.0.0.1 -v -X POST \
http://127.0.0.1:6400/api/workflows/test.yaml/copy \
-H "Content-Type: application/json" \
-d '{"new_filename":"test-copy.yaml"}'
```
9) Delete ability unit
- Method: DELETE
- URL: http://127.0.0.1:6400/api/workflows/<filename>/delete
- Example:
```
curl --noproxy 127.0.0.1 -v -X DELETE \
http://127.0.0.1:6400/api/workflows/test.yaml/delete
```
10) List local tools (function_calling)
- Method: GET
- URL: http://127.0.0.1:6400/api/tools/local
- Example:
```
curl --noproxy 127.0.0.1 -X GET http://127.0.0.1:6400/api/tools/local
```
## Tools hot updates
Local function tools are managed via the same backend (port 6400).
Endpoints:
- List local tools: `GET http://127.0.0.1:6400/api/tools/local`
- Create/overwrite local tool file: `POST http://127.0.0.1:6400/api/tools/local`
Create example:
```
curl --noproxy 127.0.0.1 -X POST http://127.0.0.1:6400/api/tools/local \
-H "Content-Type: application/json" \
-d '{
"filename": "add.py",
"content": "def add(a: int, b: int) -> dict:\n return {\"result\": a + b}\n",
"overwrite": true
}'
```
YAML example (local function_calling):
```
tooling:
- type: function
config:
tools:
- name: add
```
## Common scenarios and tool suggestions
- If the agent needs to search the web, use `web_search`.
- If the agent needs to fetch a URL's content, use `read_webpage_content`.
- If the agent needs to inspect local files, use `describe_available_files` and `list_directory`.
- If the agent needs to read a file snippet, use `read_text_file_snippet` or `read_file_segment`.
- If the agent needs to write output files, use `save_file`.
## Advanced Workflow Templates
Below are advanced YAML templates you can copy and adapt. They cover debate loops,
subgraph reuse, conditional edges, and selective payload routing.
### 1) Two-Agent Debate Loop with Judge Stop Criteria
```yaml
version: 0.4.0
vars: {}
graph:
id: debate_loop
description: Two agents debate until the judge outputs "Verdict: STOP".
is_majority_voting: false
start:
- Topic
nodes:
- id: Topic
type: passthrough
config: {}
- id: Pro
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
You are the PRO debater. Respond with arguments and counterpoints.
- id: Con
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
You are the CON debater. Respond with arguments and counterpoints.
- id: Judge
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
Evaluate the debate. Output:
Score: <0-1>
Notes: <brief>
Verdict: CONTINUE|STOP
- id: Final
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
Produce the final summary of the debate and recommendation.
edges:
- from: Topic
to: Pro
keep_message: true
- from: Topic
to: Con
keep_message: true
- from: Pro
to: Judge
- from: Con
to: Judge
- from: Judge
to: Pro
condition: need_reflection_loop
- from: Judge
to: Con
condition: need_reflection_loop
- from: Judge
to: Final
condition: should_stop_loop
```
### 2) Subgraph Reuse + Selective Payload Passing
```yaml
version: 0.4.0
vars: {}
graph:
id: parent_with_subgraph
description: Uses a subgraph and only forwards the "Summary" section.
start:
- Request
nodes:
- id: Request
type: passthrough
config: {}
- id: ResearchSubgraph
type: subgraph
config:
type: file
config:
path: "subgraphs/deep_research_executor_sub.yaml"
- id: Writer
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
Write the final report using only the provided Summary.
edges:
- from: Request
to: ResearchSubgraph
keep_message: true
- from: ResearchSubgraph
to: Writer
process:
type: regex_extract
config:
pattern: "Summary:\\s*(.*)"
group: 1
multiline: true
dotall: true
on_no_match: pass
```
### 3) Conditional Routing + Minimal Payload
```yaml
version: 0.4.0
vars: {}
graph:
id: router_with_conditions
description: Route to different nodes based on tags; forward only CONTENT block.
start:
- Router
nodes:
- id: Router
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
Decide the route. Output exactly:
ROUTE: QA|REPORT
CONTENT: <payload to forward>
- id: QA
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
Answer the question concisely.
- id: REPORT
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
Write a structured report.
edges:
- from: Router
to: QA
condition:
type: keyword
config:
any: ["ROUTE: QA"]
process:
type: regex_extract
config:
pattern: "CONTENT:\\s*(.*)"
group: 1
dotall: true
on_no_match: drop
- from: Router
to: REPORT
condition:
type: keyword
config:
any: ["ROUTE: REPORT"]
process:
type: regex_extract
config:
pattern: "CONTENT:\\s*(.*)"
group: 1
dotall: true
on_no_match: drop
```
## Tips
- Any workflow YAML can be used as a subgraph to compose more complex tasks, including existing ability units/workflows. Example:
```
- id: Research Subgraph
type: subgraph
config:
type: file
config:
path: "yaml_instance/deep_research_v1.yaml"
```
- Ensure that all generated artifacts are output to your workspace file paths whenever possible
- Ensure that all agents’ roles are concise, sophisticated, and equipped with sufficient tools

280
yaml_instance/middle_east_simulation.yaml Normal file
View File

@ -0,0 +1,280 @@
version: 0.0.1
vars: {}
graph:
id: middle_east_simulation
description: Simulate Middle East situation from multiple parties perspectives
log_level: INFO
is_majority_voting: false
start:
- Situation Brief
end:
- Final Report
nodes:
- id: Situation Brief
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are a News Analyst. Your task is to search for the latest Middle East news and create a situation brief.
[Instructions]
1. Use web_search to get the latest Middle East news
2. Focus on: US-Israel operations, Iran tensions, Lebanon/Syria situation, Gulf states response
3. Create a brief summary (2-3 paragraphs) of the current situation
[Output]
Save the situation brief to a file named situation_brief.txt
tooling:
- type: function
config:
auto_load: true
tools:
- name: web_search
- name: describe_available_files
- name: load_file
- id: US Perspective
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are playing the role of the United States Government. Analyze the current Middle East situation from the US perspective.
[Your Role]
- You represent American strategic interests
- Support Israel right to self-defense
- Want to contain Iran influence
- Protect US military personnel in the region
[Task]
Based on the situation brief, write 2-3 paragraphs analyzing:
1. Current US objectives in the Middle East
2. How US views Iran, Israel, and other actors
3. What US likely will do next
Write in first person as the US government. Be realistic and strategic.
tooling:
- type: function
config:
auto_load: true
tools:
- name: describe_available_files
- name: load_file
- id: Israel Perspective
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are playing the role of the Israeli Government. Analyze the current Middle East situation from Israel perspective.
[Your Role]
- You represent Israeli security interests
- Face existential threats from Iran and proxies
- Need to maintain military superiority
- Navigate international pressure
[Task]
Based on the situation brief, write 2-3 paragraphs analyzing:
1. Israel current security concerns
2. How Israel views Iran, Hezbollah, Hamas
3. What Israel likely will do next
Write in first person as the Israeli government. Be defensive but determined.
[Note: This is a simulation for educational purposes]
tooling:
- type: function
config:
auto_load: true
tools:
- name: describe_available_files
- name: load_file
- id: Iran Perspective
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are playing the role of the Iranian Government. Analyze the current Middle East situation from Iran perspective.
[Your Role]
- You represent Iranian regional interests
- Support Axis of Resistance (Hezbollah, Hamas, Houthis)
- Face US/Israel military pressure
- Navigate internal political challenges
[Task]
Based on the situation brief, write 2-3 paragraphs analyzing:
1. Iran regional strategy and allies
2. How Iran views US, Israel, Saudi Arabia
3. What Iran likely will do next
Write in first person as the Iranian government. Be defensive about sovereignty.
[Note: This is a simulation for educational purposes]
tooling:
- type: function
config:
auto_load: true
tools:
- name: describe_available_files
- name: load_file
- id: Arab League Perspective
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are playing the role of the Arab League / Gulf States. Analyze the current Middle East situation from the Arab world perspective.
[Your Role]
- You represent Saudi Arabia, UAE, and other Gulf states
- Compete with Iran for regional influence
- Concerned about instability and refugees
- Have normalization deals with Israel but cautious
[Task]
Based on the situation brief, write 2-3 paragraphs analyzing:
1. How Arab states view the US-Iran-Israel triangle
2. Regional stability concerns
3. What Arab states likely will do
Write in first person as the Arab League.
[Note: This is a simulation for educational purposes]
tooling:
- type: function
config:
auto_load: true
tools:
- name: describe_available_files
- name: load_file
- id: Simulation Orchestrator
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are a Political Simulation Orchestrator. Your task is to synthesize all perspectives and create an interactive simulation scenario.
[Input]
You have perspective analyses from:
- United States
- Israel
- Iran
- Arab League
[Task]
1. Read all perspective files
2. Identify key conflicts and opportunities for dialogue
3. Create 3-5 simulation scenarios showing how these parties might interact
4. For each scenario, predict possible outcomes
[Output Format]
Create a simulation scenarios document with:
- Brief: Current deadlock points
- Scenario 1: [Title] - How parties might react
- Scenario 2: [Title] - How parties might react
- Scenario 3: [Title] - How parties might react
- Risk Assessment: What could trigger major escalation
- De-escalation Path: What could lead to peace
Write in English, use clear formatting.
tooling:
- type: function
config:
auto_load: true
tools:
- name: describe_available_files
- name: load_file
- id: Final Report
type: agent
context_window: -1
config:
provider: openai
name: gpt-4o
base_url: ${BASE_URL}
api_key: ${API_KEY}
role: |-
You are a Senior Political Analyst. Create a final comprehensive report on the Middle East simulation.
[Task]
Compile all inputs into a professional analysis report:
1. Executive Summary (1 paragraph)
2. Current Situation Overview
3. Party Perspectives Summary:
- US Position
- Israel Position
- Iran Position
- Arab States Position
4. Simulation Scenarios
5. Key Takeaways
6. Outlook (next 3-6 months)
[Format]
Use markdown. Write in Chinese (Simplified). Make it comprehensive but readable.
This is for educational/analytical purposes only.
tooling:
- type: function
config:
auto_load: true
tools:
- name: describe_available_files
- name: load_file
edges:
- from: Situation Brief
to: US Perspective
condition: true
- from: Situation Brief
to: Israel Perspective
condition: true
- from: Situation Brief
to: Iran Perspective
condition: true
- from: Situation Brief
to: Arab League Perspective
condition: true
- from: US Perspective
to: Simulation Orchestrator
condition: true
- from: Israel Perspective
to: Simulation Orchestrator
condition: true
- from: Iran Perspective
to: Simulation Orchestrator
condition: true
- from: Arab League Perspective
to: Simulation Orchestrator
condition: true
- from: Simulation Orchestrator
to: Final Report
condition: true

52
yaml_instance/url_summarize.yaml Normal file
View File

@ -0,0 +1,52 @@
version: 0.4.0
vars: {}
graph:
id: url_summarize
description: 爬取URL内容并总结
is_majority_voting: false
log_level: INFO
start:
- Web Fetcher
nodes:
- id: Web Fetcher
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
你是网页内容抓取助手。你的任务是从URL提取内容。
[关键指令]
1. 必须使用 read_webpage_content 工具来抓取网页
2. URL格式: {"url": "https://example.com"}
3. 返回提取的内容
tooling:
- type: function
config:
auto_load: true
tools:
- name: web:All
- id: Content Summarizer
type: agent
config:
provider: openai
base_url: ${BASE_URL}
api_key: ${API_KEY}
name: gpt-4o
role: |
总结网页内容。
输出格式（中文）：
## 摘要
<2-3句总结>
## 关键点
- <point 1>
- <point 2>
- <point 3>
edges:
- from: Web Fetcher
to: Content Summarizer
keep_message: true