Linc AIDocumentation

Using Chat

Asking questions about your documented processes and uploaded documents — selecting sources, citations, web search, Deep Research, the prompt library, and chat history.


Chat is a natural-language layer over your documented processes, generated SOPs, analyses, and uploaded business documents — an assistant that has read every document you point it at and answers with citations back to the source material.

Querying your knowledge base

  1. Click Chat in the left sidebar and start a new chat.
  2. Select your sources in the Documents panel — expand folders and check the files or folders you want the AI to search, or leave it on Auto-select source to let the AI decide.
  3. Type your question in the text box and hit Enter. The AI searches your selected documents and responds with an answer and citations — numbered references like [1], [2] that link back to the source. Click any citation to open the document at the exact relevant page. A new chat also shows starter questions (how a process works in your documents, bottlenecks, what the last analysis found). These ask about documents and existing analysis — they do not generate an SOP or BPMN, and they do not start an Architect run. Use Scout for those deliverables and Architect for a new analysis.

Extra features

  • Web Search (globe toggle below the text box) — also search the live internet, not just your documents; useful for frameworks, best practices, and industry standards.
  • Deep Research (brain toggle) — a much more thorough analysis: the AI spends 1–5 minutes deeply analyzing your documents instead of a quick search. Great for complex questions that read across many full documents.
  • Attach images — click the paperclip to include screenshots or diagrams with your question; the AI can see them.
  • Prompt library (the "Agents" chat-bubble icon) — use or create reusable prompt templates. Prompts can include variables with {{variable}} syntax (you fill them in on use), can be Personal or shared with your Organization, and can be organized into custom color-coded categories.
  • Long conversations — long chats don't run out of room. When a conversation gets long, older messages are summarized automatically and a divider in the conversation reads "Earlier messages summarized to keep this chat going" — click View summary to see what the AI kept. Your full history stays on screen and saved. As a chat approaches its next summary, a small ring appears in the text box; hover it to see how full the context is.
  • Analysis chat suggestion (admins) — when you ask for something Architect's Analysis chat handles better — a chart, report, deck, or spreadsheet built from your analysis, a ranking or deeper analysis across your processes, or what the analysis found about processes, opportunities, gaps, or the roadmap — a box appears above the text box suggesting it. Ask in Analysis chat opens a new Analysis chat with your question already sent; the X hides the suggestion for the rest of that chat. It only appears for org admins and admins/owners of the selected Initiative, and only once that Initiative has a completed analysis.

Chat history

Your conversations are saved — find past chats in the sidebar under the nav, and rename or delete them. (The Agents and Account sidebar groups start collapsed on Chat to leave room for the history; click either label to expand it.)

Tip: Upload any business documents — ERP manuals, compliance policies, training materials, vendor contracts — into the Library. This gives Chat context on your whole business, not just the workflows you've recorded.