BuildShip

B

BuildShip

Join the community to ask questions about BuildShip and get answers from other members.

Join

❓・buildship-help

🌇・showcase-and-share

💬・general

Missing logs

I am unable to see any logs on my workflows for both test and prod! I was able to see logs before today, but can't anymore! I did rename all my workflows today, not sure if that caused it. Any help would be great 😄

how to save Current timestamp in firebase create document node

Sorry if this is a dumb question but how do you save the current timestamp to a firebase document with a data type as “timestamp” Eg. { “createdAt”: “currentTimestamp”...
Solution:
Hi @Your Majesty, you just require to use new Date() type of the editor set to Javascript.

Trigger that reacts on new files in a Supabase storage?

I need a trigger that reacts when new file is detected in a Supabase storage bucket. Is it only doable by configuring a webhook on Supabase, or does Buildship have an easier solution for it? Also: at the end of the workflow i need a node that saves the result files to another Supabase storage bucket - any tips how to do that?

Stripe webhook testing - 500 error

I have a working production Stripe webhook trigger. I want to be able to test it as well so I imported it to Sandbox. In Sandbox mode however I am getting '500 - Internal' error. I suspect it is due to the live and not the test key being in the Stripe webhook in Buildship. But then: How do I test Stripe webhook triggers without opening up a new workflow? Any experience? Thank you!...
No description

Publish suddenly doesn't work!

I am unable to publish my workflows since the last couple of hours. Publish worked fine until last night and I have hit any of my limits yet as well. I get an error message stating Failed to fetch! Any help would be great!
Solution:
Yes, the fixed is released now. Thanks for confirming.
No description

Node editor

Hi guys, I am non-technical cofounder of a startup and completely new to BuildShip. I do not have a lot of coding/developing experience and I am trying out BuildShip as a free user first to see if it can help me build an AI test workflow. I have a basic level question. When I try to look at the code for a certain node I am unable to access the Node Editor and get an error message. I appreciate any help on this. thanks....
Solution:
Thanks for confirming. Have also sent you an email regarding this, we've pushed more enhancements for node editor. Happy Building and Shipping using BuildShip.

Error response from Buildship Workflow even if success

I get an error in the Builship Workflow node, but when I check the specific workflow, I get success, it should return the data. Workflow ID that run the other workflow: 2L8uuxvp9hdAhFmtL71H Workflow ID that is being run: 9VJ9F7ECfsmpCvUhyWYh...
Solution:
Hi @Francois, from the screenshot, the BuildShip execute workflow node didn't executed successfully, as it shows a syntax error, you can check on the data if it is correct? Also, you can share your workflow via in-app support as well - https://docs.buildship.com/support-messages.
No description

Chat with your database - Assistant V2 - 'V' : 'Output too large'

I'm using the OpenAI assistant node like in the 'chat with your database' template where it queries a menu dataset. In my case the dataset is a restapi call. It kind of works. However, when I feed it too many rows it will just return 'input too large - output too large' in the logs and it will generate a random response for the assistant. I'm testing with 'how many rows are there in my dataset'.. When it is able to work through all data it will say like for example 50 which is correct. But...
Solution:
For the second question to save the data from REST API request and sue it as feed to Assistant, you can create a workflow like this to achieve the usecase.

streaming AI assistant + flutterflow & Firebase

Has anyone actually been able to implement the chat history in flutterflow?? I know that the chats are saved in Firebase, but how do you implement it in flutterflow?...

API call not working

Hi, Newbie here, I tried making an api call to openfoodfacts but I keep getting this error. It works when I make the api call in the address bar but doesn't seem to work in the buildship environment. Any help would be much appreciated. Thanks...
No description

Gemini Text Generator not working

I have been trying to get. My Gemini text generator node to work but it's not working and my API key is working because I tested it with Python. But every time I test my Gemini text generator node. I get this. Cannot read properties of undefined (reading '0')...
No description

Ship Template No API

the Ship button, it does not show me the API endpoint when I ship it, why? @Harini #ship...

ALL MY WORFLOWS ARE DOWN PLEAASE

All of my worflows are not workign anymore and the fetch to buildShip says "workflow not found" I have 50+ workflows

error message in return raw body text

Solved: I recreated the node and deleted the old one. ---- ERR_MODULE_NOT_FOUND","url":"file:///usr/src/app/bucket/builtNodes/b53faea1-f7b6-4cba-8b92-5050d0b391fd/1716541145220.cjs"},"label":"prompts","message":"Cannot find module '/usr/src/app/bucket/builtNodes/prompts/1716541145220.cjs' imported from /usr/src/app/runtime/dist/WorkflowExecution_v2/Executors/Script.js ...

Timeout via Postman

Hi Guys! I'm building complex logics where I generate 60-80 calls to OpenAI within one workflow. If I test the workflow vie BS it works fine, even when the completion takes up to 10-15minutes. However when I call the endpoint from let's say a Postman, it dies after 5 min, with t 504 gateway timeout. Although the execution still runs and completes on BS according to the logs.. I have also checked my Postman preferences, and there is no max timeout set for requests (0=never timeout) Is there any...

Global variables & webhooks

We are considering using BuildShip as a backend tool. There are however two important questions: 1. Is there a way to store global variables and trigger an action based on them? 2. Is it possible to listen to create a 'wait for webhook' in buildship? Where the flow would continue based on an incoming webhook result....

I will help you with my idea

Greetings Discord community, I am looking for opportunities to collaborate with my ideas and talents. I can increase your sales with my idea and strategy. I have a huge interest in marketing....

Whisper API - Broken

Hi guys For some reason the whisper API I use to transcribe MP3 files accessible via a URL (yes it's publicly accessible) started malfunctioning in ALL of my previously functioning workflows. It returns a text:null...

Help with building a single long form document with one workflow and multiple nodes providing input

I want to build a client or lead analysis document that provides an automated analysis on a business or prospect providing valuable data in long form analyst doc. This doc is pieced together by different specialist functions or api calls to LLM's that create parts of the document on step at a time which response data is saved somehow for the workflow to retrieve and use as it proceeds along, eventually the final document is pieced together from the different parts created and a final analysis and tweaks is done by a top tier LLM quality assurance agent/assistant. The point is through a multi step workflow run can i gather data and output from different sources during the workflow which is then pieced together at the end (prob via all parts being sent to a long content llm) to create a long form analysis document that is the final output posted as a word document somewhere or saved to google drive or emailed that was based on and created by multiple actions that occurred before and previously saved to DB data or cached by the workflow. Essentially, I just want to know if this is easy and possible to do with buildship and the workflow has a database or storage capacity or large cache, vector and normal ones to save data to (or is that not build in DB is another api call?) and that it can work with large chunks of text being passed around, any roadblocks or problems with this idea, i need to know first before i spend time learning and testing as im on the clock, just need to know if its possible and "easily done" with no potential issues. Thank you so much....

400 The requested model 'gpt-4o' cannot be used with the Assistants API in v1.

I am trying to use the openai assistant node with gpt 4o. I get this error: 400 The requested model 'gpt-4o' cannot be used with the Assistants API in v1. Follow the migration guide to upgrade to v2: https://platform.openai.com/docs/assistants/migration. Any ideas?...