Use AI to discover know-how from your documents, get cited answers, and build agent-driven automations for repeated tasks, on your terms.
Turn complex documentation into cited answers, then into repeatable workflows with templates and approvals — and connect MCP tools plus the LLM providers you already use.
Ask questions in plain language with workspace and file context. Get answers with citations back to the sources.
Search across everything you have uploaded with natural-language questions — ideal when you do not know the exact filename.
Turn answers into repeatable outcomes: triage, draft, approvals, escalation, and integrations — powered by step templates.
Create specialized agents with controlled access to tools. Draft, review, and improve skills so each agent handles repeat work more accurately and faster.
Connect Model Context Protocol servers and call their tools from workflows so your automations reach the systems you rely on.
Wire up LLM providers that fit your deployment — from local Ollama for private data to cloud models through AWS Bedrock.
Short YouTube clips per capability. Search and workflow each have a dedicated intro; approval still uses a chapter-style start on the full product tour.
Natural-language search over your uploaded documents returns a concise answer and points you to the underlying files.
Search docs →Kick off workflows that research your workspace and produce drafts — without copy-pasting between tools.
Workflows docs →Review AI-generated outputs and approve before anything is finalized — one click when you are satisfied.
Workflows docs →Pick an agent, @-mention workspace documents in the side chat, use slash shortcuts in the editor, and ask the assistant to update the open document—including MCP web search when your agent is wired for it.
Agent chat & AI agents →Start from built-in templates for agents, document tools, semantic search, MCP calls, and approval gates — then fill a short form of placeholders.
Workflows docs →Connect MCP servers for external tools in workflows, and wire up LLM providers such as self-hosted Ollama or AWS Bedrock.
MCP servers & LLMs →Search, chat, workflows, MCP, and models — common questions from teams evaluating A1KnowHow.
A1KnowHow is an intelligent knowledge base system powered by AI agents that helps you find information instantly. Instead of searching through folders and files, you can ask questions in plain English and get answers from your documents.
You turn tacit know-how into agent skills: what to check, which tools agents may use, and how to respond. Run agents in workflows, evaluate the results, and refine skills over time so each run is more accurate and repeatable.
A1KnowHow supports a wide variety of document formats including PDFs, Word documents, text files, and more. You can upload documents from various sources and they'll all be searchable through our AI-powered interface.
Yes, A1KnowHow is fully accessible on mobile devices. You can search and access your documents from anywhere, whether you're at your desk or on the go.
Yes. A1KnowHow Cloud includes a 30-day free trial with no credit card required. The trial includes a $2 usage credit and lets you create up to 100 documents. Additional usage limits apply. After the trial, A1KnowHow Cloud is $20 per month per account plus usage charges for AI tokens, embeddings, and document processing. There is no per-seat fee today; it is aimed at individuals and small teams (about 2-5 people).
We offer comprehensive documentation and support resources. You can access our help center, documentation, or contact our support team if you need assistance getting started or have any questions.
Search and agent responses are built to retrieve from your workspaces and documents first. Use the cited sources and snippets shown with answers to verify what the model used — always review critical decisions before acting.
Workflows are multi-step agentic automations you can trigger on demand. Step templates speed up building: pick a template (for example an agent step, a document tool, semantic search, an MCP tool call, or a human approval gate), fill in a short form, and refine from there.
MCP (Model Context Protocol) lets A1KnowHow connect to external tool servers. In workflows you can call MCP tools the same way you use built-in document and search steps — useful for pulling data or taking actions in other systems you already run.
Yes. You can configure LLM providers such as self-hosted Ollama for private deployments or AWS Bedrock for managed cloud models, depending on your account and setup. See documentation for provider-specific steps.
A1KnowHow supports self-hosted installations for teams that want data and models on their own infrastructure. The cloud trial and SaaS experience use the same product concepts; see the self-hosted docs for dependencies, Docker, and configuration.
No. A1KnowHow is not open source today. We may release a community edition in the future.
Deploy on your infrastructure, our managed cloud, or enterprise on-premises. Pricing follows how you host and operate A1KnowHow.
No pricing plans are currently available. Check back soon!
Start your free trial and see answers, agent chat, and workflows on your own files — no credit card required.
Start Free Trial