Understanding and Using Sub-Agents

Sub-agents are specialized AI assistants that work on specific tasks within your project. They allow parallel processing of multiple sections or tasks, significantly speeding up complex bid preparation.

What are Sub-Agents?

Sub-agents are autonomous AI workers that:

Handle specific tasks or document sections independently
Work in parallel with other sub-agents
Report back to the main agent with their results
Can be monitored and controlled individually

When to Use Sub-Agents

Sub-agents are ideal for:

Writing multiple bid sections simultaneously
Analyzing different aspects of tender requirements in parallel
Conducting research on multiple topics at once
Generating multiple document versions for comparison
Processing large volumes of information efficiently

How Sub-Agents Work

Automatic Creation

When you request a complex task, TenderB automatically creates sub-agents:

Example:

"Please write a complete tender response covering: executive summary, technical approach, team and experience, pricing, and implementation plan."

The AI will create 5 sub-agents, each handling one section.

Manual Creation

You can explicitly request sub-agents:

Example:

"Create sub-agents to analyze: 1) technical requirements, 2) commercial terms, 3) compliance requirements, and 4) evaluation criteria."

Monitoring Sub-Agents

Viewing Sub-Agent Status

In the right panel, you'll see:

List of active sub-agents
Current task for each sub-agent
Progress indicators
Completion status

Viewing Sub-Agent Output

Click "View subagent" to see:

The sub-agent's conversation thread
Work in progress
Sources and references used
Reasoning and decision-making process

Interacting with Sub-Agents

Providing Guidance

You can give instructions to specific sub-agents:

"Tell the technical approach sub-agent to emphasize our cloud-native architecture."

Requesting Revisions

Ask specific sub-agents to revise their work:

"Ask the pricing sub-agent to recalculate based on a 10% contingency."

Reviewing Progress

Check on sub-agents mid-task:

"Show me the current status of all sub-agents."

Best Practices for Sub-Agents

Clear task division: Define distinct, non-overlapping tasks for each sub-agent
Provide context: Ensure each sub-agent has access to relevant documents
Monitor progress: Check sub-agent status periodically
Review individually: Examine each sub-agent's output before combining
Maintain consistency: Provide style guidelines that apply to all sub-agents

Example Workflows

Parallel Section Writing

"Create 6 sub-agents to write the following sections simultaneously: 1) Company Overview, 2) Technical Solution, 3) Implementation Methodology, 4) Team Structure, 5) Risk Management, 6) Quality Assurance. Each section should be 2-3 pages."

Multi-Aspect Analysis

"Deploy sub-agents to analyze: 1) Technical feasibility, 2) Commercial viability, 3) Resource requirements, 4) Risk factors, 5) Compliance issues. Synthesize findings into a Go/No-Go recommendation."

Comparative Research

"Create sub-agents to research and compare: 1) Our solution vs. Competitor A, 2) Our solution vs. Competitor B, 3) Our solution vs. Competitor C. Focus on features, pricing, and customer satisfaction."