Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| EXA_API_KEY | No | Your Exa API key required if the Researcher agent is used and needs Exa | |
| DEEPSEEK_API_KEY | Yes | Your DeepSeek API key required for the LLM used by Agno agents/team | |
| DEEPSEEK_BASE_URL | No | Optional: Custom DeepSeek endpoint URL if using a custom endpoint |
Schema
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| sequential-thinking | Enhanced starter prompt for sequential thinking with better formatting. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| sequentialthinking | Advanced sequential thinking tool with multi-agent coordination. Processes thoughts through a specialized team of AI agents that coordinate to provide comprehensive analysis, planning, research, critique, and synthesis. Args: thought: Content of the thinking step (required) thought_number: Sequence number starting from 1 (≥1) total_thoughts: Estimated total thoughts required (≥5) next_needed: Whether another thought step follows this one is_revision: Whether this thought revises a previous thought revises_thought: Thought number being revised (requires is_revision=True) branch_from: Thought number to branch from for alternative exploration branch_id: Unique identifier for the branch (required if branch_from set) needs_more: Whether more thoughts are needed beyond the initial estimate Returns: Synthesized response from the multi-agent team with guidance for next steps Raises: ProcessingError: When thought processing fails ValidationError: When input validation fails RuntimeError: When server state is invalid |