Most manufacturing teams have access to powerful AI tools, but those tools don’t know anything about their specific process. They’re general-purpose, and while useful for drafting text or answering broad questions, are disconnected from the actual system you’re trying to improve.
The AI Assistant in Threaded is different. It operates with full context of your organization including your Work Instructions library, Map, parts and tools catalogs, issues, and org info. This turns AI into a collaborator that understands what you build and how you build it, not just a chatbot you have to explain everything to from scratch.
#Agentic Capabilities
AI in Threaded can give you advice and analysis of your operation, but it doesn’t stop there — it can also act. It can execute changes directly in the platform on your behalf, not just telling you what to do. That includes updating work instructions, creating and assigning issues, and making other changes to your process. Connecting Threaded MCP in the chat sidebar expands what the assistant can read and update on your behalf.
When the AI proposes a change, it presents a summary of what it intends to do and waits for your confirmation before executing. You review the proposed changes and click Execute to apply them, or dismiss them if you want to take a different approach. The AI acts only on your approval, and you remain in control throughout.
This human-in-the-loop approach means you can delegate repetitive or time-consuming tasks to the AI — like updating a step across multiple procedures, or creating a set of issues from an analysis — while maintaining full visibility and control over what actually changes in your system.
#Auto-accept tools (bolt)
By default, every tool call waits for you to click Execute. For longer workflows — especially multi-step report authoring or bulk updates — you can turn on auto-accept tools so approved tools run without a click each time.
- Open the AI Assistant sidebar.
- Click the bolt icon next to the model selector / Threaded MCP controls.
- When auto-accept is on, the bolt stays highlighted and the tooltip reads Auto-accept tools is on. Click again to turn it off.
Auto-accept is saved per conversation. Clarifying questions the assistant asks you (selectable options) still require your answer before related mutations continue — those prompts are never auto-accepted.
Use auto-accept when you trust the current thread and want speed; leave it off when you want to review every change before it runs.
#Accessing the AI Assistant
Click the AI Assistant icon in the top navigation bar to open the chat sidebar, which is accessible from anywhere in Threaded. You can close the sidebar by clicking the close arrow, or open it as a full page with the expand button.
Each conversation starts fresh by default, but the AI has access to memory from previous conversations and can reference past discussions when relevant. To start a new chat, click the + at the top of the AI assistant sidebar, or to access a prior chat, click the history icon right next to it.
#How Context Works
The quality of the AI’s responses depends directly on the context it has access to. A general-purpose AI has to work from whatever you tell it in the moment, which means you spend as much time explaining your process as actually improving it. Threaded solves this by building context into the conversation from the start, and giving you tools to add more when you need it. Threaded provides context to the AI in several ways:
#Automatic Context
Every conversation automatically includes a high-level summary of your organization: published Work Instructions, Map process nodes, parts and tools, issues, and the Organization Info you’ve configured under Admin. The AI also has tools to search and navigate this structure to target specific parts of your process in response to your prompts. The more information you put into Threaded, the more relevant and specific the AI’s responses will be to help you engineer and scale your process.
#Adding Context with @mentions
Type @ in the chat to pin specific objects into the conversation as focused context. The AI has direct access to anything you @mention and can analyze, reference, and reason about it specifically. You can @mention:
- Work Instructions and process nodes
- Parts and tools
- Issues
- Media files
- Organization members
- Previous AI conversations
For a full breakdown of what can be @mentioned and where, see Search and Mentions.
#Attaching Files
Click the paperclip icon below the chat window to attach external files—documents, spreadsheets, images, PDFs, and more. This is useful for bringing in reference material, supplier data, quality reports, or any other context that lives outside of Threaded. Files brought into context through chat can also be saved for future reference in the media library by clicking the “Save to Organization Media” icon in the chip for the attached file.
#Conversation History
By default, the AI includes memory from previous conversations as context, so it can reference prior discussions and build on past analysis. To start a fully clean session without previous conversation context, click the conversation history icon next to the model selector to disable it in the event that you want to start completely fresh.
#Guided Workflows
Beyond freeform chat, the AI Assistant can run structured workflows that walk you through a multi-step task. Ask for the workflow by name, or start from the relevant page in the app:
- Gather Organization Information — collect and save Organization Info (industry, locations, priorities, challenges, and related context) so every future conversation has better grounding. You can also start this from Admin > Organization Info.
- Review Work Instructions — a conversational review of completeness and quality for one or more work instructions. This is collaborative guidance in chat, not an automated Process CI run. For scored checks before publish, use the Work Instruction Review tab.
- Author a Report — create or update Manufacturing Reports (planning document + HTML shell) in the AI sidebar.
#Model Selection
Threaded gives you access to multiple frontier AI models, all in the same platform. Different models may perform better on different types of tasks, so if one isn’t giving you what you need, switching is a quick way to get a different perspective. You can switch models using the model selector below the chat window.
#Tips for Getting the Most Out of the AI
- Add context before asking — @mention the specific procedures, nodes, or parts you want to analyze before asking your question. AI performs significantly better when it has focused context rather than working from a broad org-level summary alone.
- Be specific about your goal — “Analyze my Powder Coat operation for bottlenecks” will get you a more useful response than “analyze my process.”
- Use it as a collaborator, not just a search tool — ask follow-up questions, push back on recommendations, and ask the AI to explain its reasoning. The best results come from a back-and-forth conversation.
- Let it act — if the AI suggests a change and you agree with it, use the Execute workflow rather than making the change manually. It’s faster and keeps the conversation in context.
- Use the bolt for trusted multi-step work — turn on auto-accept tools when you want a longer tool sequence to run without clicking Execute on every step; turn it off when you need to review each action.
- Start a new conversation when shifting focus — the AI works best when a conversation is focused on a single goal or topic. When you move to a new problem, start a new chat.
#What the AI Can Help With
The AI Assistant is useful across the full range of work in Threaded:
- Work Instructions — draft new procedures, improve existing steps, check for clarity and completeness in conversation, and make bulk updates across your library. For automated completeness and quality scoring before publish, use Process CI on the Work Instruction Review tab.
- Process Analysis — identify constraints and bottlenecks, analyze line balance, compare Process Plan vs Actuals, and estimate the impact of proposed changes
- Continuous Improvement — surface improvement opportunities, help prioritize issues, and connect floor feedback to process changes
- Parts and Tools — analyze usage, identify supply chain dependencies, and help populate or clean up your parts and tools tables
- Scenario and Project Planning — ask how to double production in a given timeframe, or to provide a plan to launch production of a new product. It will outline a plan and create issues for users to accomplish to complete the objective.
- Manufacturing Reports — on a report’s Report details page, plan a draft report, run read-only SQL against a report SQLite database, and scaffold an HTML dashboard shell. See Manufacturing Reports and the report prompts in AI Assistant Use Cases.
- General Manufacturing Knowledge — ask about lean principles, industry standards, technical writing, or any manufacturing concept and get answers in the context of your specific process
For detailed examples and prompts organized by use case, see AI Assistant Use Cases.