SensAI DocsYour AI Agents
SensAI comes with six specialized AI agents, each designed for a specific type of task.
What Are AI Agents?
AI Agents in SensAI are pre-configured AI assistants built for specific domains. Instead of a generic chatbot, each agent arrives with deep expertise in its area — the right instructions, context, and tooling baked in so you can get useful results immediately.
Pre-Configured Expertise
Each agent has been trained and prompted with domain-specific knowledge so it understands the vocabulary, workflows, and goals of its area without any setup from you. This means you skip the back-and-forth of explaining context and get directly to useful output.
No Prompting Required
You don't need to write prompts or configure anything. Just open the agent and type in plain language, the same way you'd describe a task to a colleague. The agent handles the translation from your intent to a well-structured AI interaction.
Available Agents
| Agent | Best For |
|---|---|
| LLM Farm | Comparing, routing, and benchmarking multiple AI models |
| Knowledge Agent | Searching and Q&A across your organization's documents |
| FSD Generator | Turning rough ideas into detailed functional specifications |
| Report Summarizer | Distilling long reports into concise, actionable briefs |
| Presales Agent | Drafting RFP responses, proposals, and competitive analysis |
| Debug Visualizer | Visualizing stack traces, data flows, and bug timelines |
How Agents Work
When you open an agent, SensAI has already configured the underlying AI with specialized instructions, relevant context, and tools suited to that domain. You simply describe what you need in natural language and the agent takes care of the rest.
Every agent is built around three ideas:
Specialized Instructions
The agent understands the vocabulary, patterns, and goals of its domain without you having to explain them. Each agent carries a curated system prompt and contextual knowledge that shapes every response to be relevant to its specific use case.
Recommended Models
Each agent is paired with a model (or set of models) that performs best for its type of task. For example, a code-heavy agent like Debug Visualizer is matched to models strong at reasoning about execution flows, while Report Summarizer favors models optimized for long-context comprehension.
Unique Capabilities
Beyond conversation, agents can ingest documents, generate structured outputs, visualize data, and more. These capabilities are surfaced in each agent's interface and tailored to the workflows that matter most for that domain.
You can switch between agents at any time from the sidebar. Your conversations are saved separately for each agent.