openbmb / proagent Goto Github PK
View Code? Open in Web Editor NEWAn LLM-based Agent for the New Automation Paradigm - Agentic Process Automation
License: Apache License 2.0
An LLM-based Agent for the New Automation Paradigm - Agentic Process Automation
License: Apache License 2.0
While working with [Name of the library or project, e.g., n8nPythonNode], I encountered a TypeError when trying to use generic type annotations in Python 3.8. This issue occurs in the function run_node when using the syntax list[dict].
Use Python 3.8.
Define a function with the following signature:
def run_node(node: n8nPythonNode, input_data: list[dict] = [{}]) -> tuple[str, str]:
# function body
Expected Behavior
The function should accept a list of dictionaries as an input parameter without any type errors.
A TypeError is raised: 'type' object is not subscriptable.
For Python versions prior to 3.9, it might be necessary to import List and Dict from the typing module for generic type annotations. The function signature can be modified as follows:
from typing import List, Dict, Tuple
def run_node(node: n8nPythonNode, input_data: List[Dict] = [{}]) -> Tuple[str, str]:
# function body
This issue seems to be specific to Python versions below 3.9, where the newer syntax for generic type annotations is not supported.
Hi there,
We believe AI+Automation is the way to go, so we developed a visual scripting language called 鸿图,
https://developer.kaihong.com/docs/hongtu/
We also had some idea for LLM usage
https://developer.kaihong.com/docs/hongtu/connected/ai/llm.html
contact us if you are intrested
ModuleNotFoundError: No module named 'requests' 等等
I've been exploring your impressive accomplishments in RPA . I am trying to setup the ProAgent code in the development environment. n8n-compiler in your code base has used an older version. I'm trying to run the code using n8n 1.0.1 Version. Changed the WEBHOOK_URL . Though code could able to generate the workflow I'm facing issues in the n8n as below . I couldn't able to get the workflow nodes in n8n.Could you please share the details of the n8n version that supports your code base?
System Config:
Windows 10
Node Version: v18.20.3
python: 3.10.4
OpenAI: Azure OpenAI 1.31.1
n8n: 1.0.1
2024-07-09T11:10:31.424Z [Rudder] error: got error while attempting send for 3 times, dropping 2 events
2024-07-09T23:16:12.644Z [Rudder] error: got error while attempting send for 3 times, dropping 1 events
2024-07-10T05:16:13.013Z [Rudder] error: got error while attempting send for 3 times, dropping 1 events
2024-07-10T11:15:59.095Z [Rudder] error: got error while attempting send for 3 times, dropping 1 events
2024-07-10T17:15:59.102Z [Rudder] error: got error while attempting send for 3 times, dropping 1 events
Hi there,
Thanks for sharing the repo.
Please add a LICENSE
file to the repo.
Thank you.
Do you support to interact with n8n inside docker file, where can I find the doc related to it? Thanks in advance.
Give your next action together with thought, plans and criticisim:
get a llm_server response from Record
{'prompt_tokens': 5172, 'completion_tokens': 404, 'total_tokens': 5576}
content: Thoughts:
- 由于Google Sheets的documentId和sheetName都是以URL的形式给出的,所以我们需要在params中设置"documentId"和"sheetName"为对应的URL。
- 由于Slack Channel已经指定为#general,所以我们需要在params中设置"channelId"为#general。
- 对于Gmail和AI,由于没有给出具体的参数,我们可以先不设置params。
Plan:
- 实现读取Google Sheets数据的action。
- 实现发送Slack消息的action。
Criticisms:
- 当前计划看起来可行,没有明显的问题。
Thought: 根据Google Sheets的URL设置了documentId和sheetName的参数。
Plan:- 实现发送Slack消息的action。
Criticism: 目前计划没有问题。
Tool: function_rewrite_params
Tool Input: {'function_name': 'action_0', 'params': '{"documentId": {"mode": "url", "value": "https://docs.google.com/spreadsheets/d/1_B038J1f3VW94z179OagFwEtnr3k5mTyXKBTPI2Fw-U/edit#gid=1759497495"}, "sheetName": {"mode": "url", "value": "https://docs.google.com/spreadsheets/d/1_B038J1f3VW94z179OagFwEtnr3k5mTyXKBTPI2Fw-U/edit#gid=1759497495"}}', 'comments': '从Google Sheets中读取业务流数据,其中documentId和sheetName都是通过URL给出的。', 'TODO': ['测试此功能']}
Tool Output: {"result": ["params["documentId"] parsed with "mode"=url", "params["sheetName"] parsed with "mode"=url"], "status": "ToolCallSuccess"}
Tool Call Status: ToolCallSuccess
Now the codes looks like this:
就像上面这些
如果使用GPT4之外的LLMs,处理一般复杂的任务,能有一个不错的效果,
那真是令人惊艳和兴奋。
I've been diving into your awesome work on RPA and I’m really impressed by what you’ve achieved. 😊
I noticed you mentioned a self-constructed evaluation dataset in your paper and I'm really interested about it. I believe it could be a gold mine for folks like me who are interested in RPA research.
So, I was wondering, is this dataset available somewhere? Or are there any plans to release it to the public?
Thank you for your reply.
since I set WEBHOOK_URL=https://n8n.x-agent.net/redirect/http%3A%2F%2Flocalhost%3A5678/n8n, and n8n generates OAUTH Redirect URL: https://n8n.x-agent.net/redirect/http%3A%2F%2Flocalhost%3A5678/n8n/rest/oauth2-credential/callback, and I put it into GitHub OAUTH Application registration for a try.
But once I click connection
, it raises 522 error.
It seems a Host Error in n8n.x-agent.net.
What can I do? Is there an other way to finish connect?
Is the n8n redirection service running? When to open source redirect code?
请问n8n是否类似集简云(https://www.jijyun.cn/)..如果是,那么是否可以支持集简云以便在国内使用?谢谢
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.