-
Fin Chatbot is now live across supported Glean sites!
🎉 Getting answers to your Glean questions is now much easier with our new AI-powered support chatbot! You can find it in the bottom right corner of all the following supported sites: The Gleaniverse (Here!) Help Center Developer Site Support Portal 🎫 If you have access to create tickets in the Support Portal (Zendesk -…
-
Re-Index Deleted Document
I have admin permissions within our Glean resource. We are in the process of adding a custom data source through the indexing API. We were able to easily index every file necessary, however in order to get a better understanding of the capabilities within the python SDK I tried seeing if I could delete a document and…
-
Glean Accessing file in S3
Our team is working on leveraging Glean for a critical part of our test case generation process, which involves analyzing design assets (Figma). We are attempting to have Glean read a PNG image file that is hosted on an Amazon S3 bucket. The link to this file is generated via the Figma REST API's export function. The…
-
Generate Slide Deck
I have an internal user who would like to be able to generate a PowerPoint slide deck based on a series of prompts and responses. Ideally a script for presenting the deck would also be generated
-
Connector issues are inevitable
I’ve been supporting Glean at my company for well over a year now. Since we first implemented it in May 2024, Glean has evolved significantly, new features have been introduced, and many existing ones have become much more stable and robust. However, one thing has remained unchanged: the frequency of connector and indexing…
-
[Feature Request] PagerDuty Connector Enhancements - "on-call" names
Internal tracking is FR-4422. Glean currently does not provide real-time “on-call now” names directly in Glean Chat; instead, you’ll need to use the provided links to view the most current on-call information. This limits the types of agents that can be built for this use case.
-
Re-use Sub-Agents within an Agent
There are times where you have crafted a sub-agent that fits along different branches within your agent's logic, but currently there's no way to copy that sub-agent and re-use it. Is this a feature customers would like to see? Comment and react for increased visibility
-
Issue with “Copy message” formatting and lack of direct support channel
Hi team, When I click “Copy message” from the assistant and paste it into Slack, Google Docs, or elsewhere, the text is transferred with markdown formatting (### headings, * lists, link syntax). Since Slack and Google Docs don’t fully support this, the formatting often breaks — and in many cases, the links don’t appear at…
-
Issues with Glean Extension while using it with Emails
There are few issues noted with the Glean Extension, especially with the Translate feature Often times, when you translate and press "replace", it still performs the "Insert" operation. If I translate to Spanish, I always get an additional line at the end of the body, which is something like ÂżTe gustarĂa adaptar el tono o…
-
Batch Processing While Iterating Through Sheet Rows
I am trying to build an agent that iterates through rows to answer a question based on the context provided in each row. I have an agent set up successfully, but it will only produce a portion of results instead of all results (there should be around 90). It will display a message such as: "The above pattern continues for…
-
Is possible to define Chat Name when using SDK?
I'm using Glean SDK renderChat() to render an Agent and it works. I'm able to send initial messages + store the chatId. But I notice that all chats are named with the Agent name. I would like to define it dynamically via code while rendering the chat or when it's created. Is it possible?
