DeepSeek Harness: The Plugin-Based Runtime for AI Agents
DeepSeek recently released DeepSeek Harness (dsh). It is not a new AI model. It is an open-source runtime layer that sits around a model to manage tools, skills, sessions, storage, planning, sub-agents, and the user interface. Released under the MIT License, it gained over 160,000 GitHub stars in just six days.
So, what makes it special? It comes down to one main idea.
The Core Idea: Everything is a Plugin
DeepSeek Harness is built on the Cordis microkernel. The core is tiny. It only handles loading plugins and resolving dependencies. Everything else is a separate plugin you can swap out.
This includes:
- The AI model connector
- The tools and skills
- The session memory
- The execution sandbox
- The agent's main loop
- The user interface
You do not need to rewrite the runtime. You just change the configuration to swap storage, add tools, or build custom modes.
The trade-off? You have more compatibility boundaries to test. Plugin versions, permissions, and shared services all need to work together.
How It Works: The Trajectory View
DeepSeek Harness records everything the model sees into an append-only session log. This includes system instructions, reasoning, tool calls, results, sub-agent dispatches, and injected context.
The Trajectory view lets you inspect these records by source. You can resume, fork, search, and replay from the same event stream. If the agent changes the wrong file or stops too early, you can see exactly what happened. It does not guarantee correctness, but it makes runs much easier to reproduce.
Four Modes for Different Tasks
The runtime comes with four preset modes:
- Standard: Everyday coding-agent work. Full tools, skills, planning, and workflows.
- Code: Tool-heavy, multi-step tasks. The model can combine tool operations using TypeScript.
- Minimal: Model and harness evaluation. Only a persistent shell and file editor.
- Creator: Harness and plugin development. Adds runtime inspection and plugin experiments.
For repository work, start with Standard. To observe model behavior alone, use Minimal. To build the harness itself, use Creator.
How It Compares to Other Tools
These tools work at different layers:
| Option | Best When | Main Trade-off |
|---|---|---|
| DeepSeek Chat/API | You just need a reply | No built-in tools, memory, or execution environment |
| DeepSeek Harness | You want an open, inspectable, plugin-based runtime | Still a preview, changes fast, requires hands-on configuration |
| Claude Code | You already use Claude's native coding workflow | Tied to Claude's model and workflow ecosystem |
| OpenCode | You want a terminal-first agent with provider freedom | Integration depth varies by provider and configuration |
The Main Caveat
DeepSeek Harness is still a developer preview. DeepSeek has clearly stated that there will be compatibility-breaking changes in the future. If you use it as fixed infrastructure, lock your package versions, keep configuration in version control, and re-run a small test suite before updating.
Architecture alone does not prove it is better than mature coding agents. Reliability must be measured with your own repository, tools, and models.
For developers who want to inspect, customize, and own their agent runtime, it is a project worth watching right now.
