[GH-ISSUE #5353] [BUG]: Unsupported provider [object Object] while web scraping #5038

Closed
opened 2026-06-05 14:51:39 -04:00 by yindo · 2 comments
Owner

Originally created by @galionsfigur on GitHub (Apr 4, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5353

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

When instructing an agent to scrape and summarize a webpage in the Desktop App, a failure occurs if the scraped content exceeds the model's context limit. The agent correctly identifies the context limit issue and attempts a fallback summarization, but then crashes with an Unsupported provider [object Object] error.

Tool Call: web-scraping({"url":"https://example.com"})
@agent is executing `web-scraping` tool { "url": "https://example.com" }
@agent: Scraping the content of https://example.com
@agent: This page's content exceeds the model's context limit. Summarizing it right now.
@agent: Web Scraping Error: Unsupported provider [object Object] for this task.

Are there known steps to reproduce?

  • Open the AnythingLLM Desktop App.
  • Provide a prompt instructing the agent to scrape and summarize a webpage (e.g https://example.com).
  • The agent initiates the web-scraping tool.
  • Observe the error output in the chat/logs.
Originally created by @galionsfigur on GitHub (Apr 4, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5353 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? When instructing an agent to scrape and summarize a webpage in the Desktop App, a failure occurs if the scraped content exceeds the model's context limit. The agent correctly identifies the context limit issue and attempts a fallback summarization, but then crashes with an Unsupported provider [object Object] error. ``` Tool Call: web-scraping({"url":"https://example.com"}) @agent is executing `web-scraping` tool { "url": "https://example.com" } @agent: Scraping the content of https://example.com @agent: This page's content exceeds the model's context limit. Summarizing it right now. @agent: Web Scraping Error: Unsupported provider [object Object] for this task. ``` ### Are there known steps to reproduce? - Open the AnythingLLM Desktop App. - Provide a prompt instructing the agent to scrape and summarize a webpage (e.g https://example.com). - The agent initiates the web-scraping tool. - Observe the error output in the chat/logs.
yindo added the possible bugneeds info / can't replicate labels 2026-06-05 14:51:39 -04:00
yindo closed this issue 2026-06-05 14:51:39 -04:00
Author
Owner

@timothycarambat commented on GitHub (Apr 5, 2026):

This is an issue with the summarize subtask - not web scraping. What LLM provider are you using?

<!-- gh-comment-id:4189143334 --> @timothycarambat commented on GitHub (Apr 5, 2026): This is an issue with the summarize subtask - not web scraping. What LLM provider are you using?
Author
Owner

@Akhil373 commented on GitHub (Apr 21, 2026):

I was able to reproduce this issue with the Groq provider, the actual error was a Groq 429 TPM rate limit in the summarize subtask. I've opened a PR #5476 fixing the error logging in both files so the real error surfaces properly.

<!-- gh-comment-id:4289776046 --> @Akhil373 commented on GitHub (Apr 21, 2026): I was able to reproduce this issue with the Groq provider, the actual error was a Groq 429 TPM rate limit in the summarize subtask. I've opened a PR #5476 fixing the error logging in both files so the real error surfaces properly.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#5038