Solutions

Solutions

Solutions are high-level containers that group related agents, knowledge bases, and content sources into a cohesive unit. They provide a single place to manage all the resources that work together to solve a particular business problem.

What are Solutions?

A solution brings together:

  • Agents — AI workflows that process information and generate outputs
  • Knowledge Bases — Collections of indexed content for semantic retrieval
  • Content Sources — RSS feeds, websites, and files that supply raw data

By grouping these resources under one solution, you get a clear overview of how they relate and can manage them as a unit rather than individually.

Creating a Solution

Via the App

  1. Navigate to Solutions in your account
  2. Click Create Solution
  3. Provide a name and optional description
  4. Save the solution
  5. Link existing agents, knowledge bases, and content sources from the solution detail page

Via the API

Create and manage solutions programmatically using the REST API:

curl -X POST https://api.seclai.com/solutions \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "My Solution", "description": "Groups my production resources"}'

See API Examples for complete code samples including linking resources and using AI assistants.

Linking Resources

After creating a solution you can link and unlink agents, knowledge bases, and content sources at any time. Linking is non-destructive — the underlying resources remain independent and can belong to multiple solutions simultaneously.

ActionDescription
Link AgentsAssociate one or more agents with the solution
Unlink AgentsRemove the association without deleting the agent
Link Knowledge BasesAssociate knowledge bases with the solution
Unlink Knowledge BasesRemove the association without deleting the KB
Link Content SourcesAssociate content sources with the solution
Unlink Content SourcesRemove the association without deleting the source

AI Assistants

Solutions include three built-in AI assistants that help you configure and optimise your setup. All assistants follow a propose-then-accept pattern: you describe what you want in natural language, the assistant generates a plan of specific actions, and you review and accept or decline. The assistant never makes changes without your approval.

Source AI Assistant

Analyses your solution and suggests new content sources to add.

Access: Solution detail page → Sources tab → AI Assistant.

Describe the kind of data you need and the assistant will recommend RSS feeds, websites, or other sources and can create them for you once you accept.

Example prompts:

  • "Find RSS feeds for AI and machine learning news"
  • "Add the company blog at blog.example.com as a source"
  • "I need product documentation sources for our developer portal"

MCP tool: generate_source_plan

Knowledge Base AI Assistant

Helps organise content into knowledge bases.

Access: Solution detail page → Knowledge Bases tab → AI Assistant.

It analyses your existing sources and suggests logical groupings, creates new knowledge bases, and links them to your solution.

Example prompts:

  • "Group my sources into knowledge bases by topic"
  • "Create a knowledge base for all my product documentation sources"
  • "I have 15 news sources — should they be one KB or several?"

MCP tool: generate_kb_plan

Solution AI Assistant

Provides holistic recommendations for your entire solution — from creating new resources to restructuring existing ones.

Access: Solution detail page → AI Assistant button.

What it can do:

  • Review your current setup and suggest improvements
  • Add new resources (sources, KBs, agents) to fill gaps
  • Restructure knowledge bases for better retrieval quality
  • Recommend configuration changes across multiple resources

Example prompts:

  • "Review my solution and suggest improvements"
  • "Add a customer FAQ knowledge base with relevant sources and an agent to answer questions"
  • "Reorganise my sources — group product docs into one KB and news into another"

MCP tools: generate_solution_plan / accept_solution_plan / decline_solution_plan

Tips: Use the Solution Assistant when you need cross-resource changes. For focused single-resource tasks, the Source or KB assistant is faster. For general best practices on getting the most from all AI assistants, see AI Assistant Best Practices.

MCP Tools

All solution operations are also available as MCP tools, allowing AI coding assistants and other MCP-compatible clients to manage solutions on your behalf. See the MCP Server documentation for the full list of solution tools.

Exporting

Export a solution's full configuration — including linked agents, knowledge bases, content sources, and memory banks — as a downloadable JSON file. Optionally include the content from all linked knowledge bases and memory banks.

  • UI: Open the solution detail page and click the Export button. If content inclusion is available, you'll see an estimate of the export size before confirming.
  • API: POST /authenticated/resource-exports with resource_type: "solution" and the solution ID.
  • MCP: Use the create_resource_export tool with resource_type: "solution".

See Export Formats → Solution for the full file schema and available options.

Next Steps

  • Agents — Learn about the agents you can link to solutions
  • Knowledge Bases — Understand how knowledge bases index content
  • Content Sources — Explore the types of content sources available
  • API Examples — Complete API code samples for solutions
  • MCP Server — Use MCP tools to manage solutions programmatically