Repository Issue Activity (beta)

#2100 Prefetched sources (requires_scraping=False) never enter visited_urls, so they don't dedupe and drop out of References

#2098 Bug: GoogleSearch YouTube filtering is case-sensitive

#2097 Bug: failed chat search reuses metadata from the previous successful search

#2096 Bug: MCP transport autodetection treats uppercase URL schemes as stdio

#2095 Bug: parse_search_queries_response accepts whitespace-only required fields

#2094 Bug: an all-blank retrievers header bypasses the default retriever

#2093 Bug: complementary web context is split into space-separated characters

#2089 OpenAI-compat LLM via vllm_openai so OPENAI_BASE_URL does not steal embeddings

#1526 No embedding data received

#2057 feat(scraper): reduce LLM context tokens for documentation pages

#2056 Static scan: 2 manifest/implementation gap(s)

#1800 Add Obscura as a scraper backend

#1993 Integration with open-research-lite for 85% Web Ingestion Token Reduction

#2077 WebBaseLoaderScraper discards extracted content when image/title enrichment fetch fails

#1764 Research is getting stucked every other time while giving no meaningful logs / ui hints indicating what it was attempting to do

#1534 Running research on local documents doesn't work.

#1483 GPTR Randomly Fails to Generate a Report.

#1468 Connect with Langfuse

#1575 Can't out put PDF and more

#1767 Language not support well

#1684 Feature request: anybrowse MCP integration for Cloudflare-protected scraping

#1617 Latest version cannot final export report documents

#1586 WebSocket error, fixable with reload

#1580 Cloudflare MITM included in the self-hosted codebase

#1523 MCP server configurations not found when agent asks follow-up questions