DeepSeek Harness: Building and Extending AI Coding Agents Training Course

Overview of DeepSeek Harness

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

Setup and Initial Execution

• Node.js prerequisites and package installation
• Launching 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

• Selecting and switching between workspaces
• Navigating the session composer and conversation flow
• Reading, modifying, and generating files

Executing Coding Tasks

• Summarizing repositories and navigating codebases
• Running commands and delegating tasks
• Managing and maintaining task plans

Permissions and Sandbox Environment

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

Native Tools and Integrations

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

Sub-agents and Task Delegation

• Creating and forking sub-agents
• Product providers for integrating other coding agents
• Distributing work across sessions

The Cordis Plugin Model

• Plugins defined as services, events, and effects
• Shared context and lifecycle management
• The rationale behind the absence of a privileged core

Profiles, Bundles, and Configuration

• Layering profiles and bundles
• Modifying configuration via cordis.patch.yml
• Reviewing the boot process tree

Core Subsystems

• The append-only session log structure
• Assembly of system prompts and tool schemas
• Understanding the agent loop and turn flow

Events and Extension Points

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

Plugin Development

• Registering services under specific context keys
• Incorporating tools and model adapters
• Releasing and discovering plugins

SDK, Headless Mode, and Automation

• Controlling the harness from external processes
• Utilizing TypeScript and Python SDKs
• Implementing CLI and headless automation