DeepSeek Harness: Building and Extending AI Coding Agents Training Course

Introduction to DeepSeek Harness

• Defining DeepSeek Harness and its role in the ecosystem
• The "everything is a plugin" design principle
• Overview of Web UI, CLI, and headless modes

Setup and Initial Launch

• Node.js prerequisites and package installation
• Starting the Web UI using npx
• Compiling and running from source code

Model Setup

• Configuring the DeepSeek API key
• Selecting and switching between models
• Integrating OpenAI-compatible endpoints

Workspaces and Sessions

• Managing and switching between workspaces
• Using the session composer and dialogue flow
• File operations: reading, editing, and generation

Executing Coding Tasks

• Repository summarization and code navigation
• Command execution and task delegation
• Maintaining and updating task plans

Permissions and Sandbox Environment

• Defining approval and permission policies
• Restricting file and command access
• Implementing local and remote sandbox backends

Native Tools and Integrations

• Filesystem, terminal, and shell utilities
• Support for Model Context Protocol
• Integration with Language Server Protocol

Subagents and Delegation Strategies

• Creating and forking subagents
• Product providers for external coding agents
• Distributing work across a session

The Cordis Plugin Architecture

• Plugins functioning as services, events, and effects
• Understanding shared context and lifecycle
• The rationale behind the absence of a privileged core

Profiles, Bundles, and Settings

• Layering profiles and bundles
• Modifying configuration via cordis.patch.yml
• Examining the initialized configuration tree

Core System Components

• The append-only session logging mechanism
• Assembly of system prompts and tool schemas
• Flow of the agent loop and turns

Events and Extension Capabilities

• Handling session, agent, and capability events
• Managing waterfall and interceptor events
• Aligning behaviors with specific extension points

Plugin Development

• Registering services under context keys
• Incorporating new tools and model adapters
• Distributing and locating plugins

SDK, Headless Mode, and Automation

• Controlling the harness from external processes
• Utilizing TypeScript and Python SDKs
• Automating via CLI and headless operations