# December 31: Support for tool calling in prompt() mutation, user-based memory, fact extraction ### New Features * 💡 Graphlit now supports **tool calling in the `prompt` mutation**. You can pass a `tools` argument to enable function calling within single-turn prompts, with tool call responses properly returned in the completion. * 💡 We have introduced **Facts** as a first-class entity type. Facts are extracted assertions from content (including memories) with entity mentions, text spans, and temporal validity (`validAt`, `invalidAt`). Facts support evidence linking back to source content and can be queried independently via the `facts` query. * We have added an **Agentic specification type** to differentiate specifications used for `promptConversation` vs `streamAgent`, enabling better conversation processing within agentic contexts. * We have added support for **Slack OAuth connector**, allowing user-based authentication in addition to bot tokens. ### Bugs Fixed * GPLA-4795: Fixed content enrichment errors caused by invalid URI parameters * GPLA-4797: Fixed fact extraction failures from text documents with null values * GPLA-4799: Fixed pagination errors when skip parameter exceeds 100,000 * GPLA-4794: Fixed null reference errors when parsing certain Slack messages --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://changelog.graphlit.dev/december-2025/readme-14.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.