deepseek-better-harness
Run the following command in DeepSeek Harness:
dsh plugin install guyuefangyuanl/deepseek-better-harness
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, run the command dsh plugin install guyuefangyuanl/deepseek-better-harness, and ensure you use the complete GitHub source code URL https://github.com/guyuefangyuanl/deepseek-better-harness.
About this plugin
Traditional agent review tools are often too broad, leading to false positives or unnecessary dependencies. deepseek-better-harness reimagines an agent review project by separating concerns into three independent skills. It packages this complexity into a single DSH bundle for the DeepSeek Harness.
The plugin offers three distinct skills: session-evidence to detect repetitive work and friction in sessions; project-readiness to assess whether a repo provides a clear change path; and agent-assets to verify the discoverability and integrity of skills and configurations. Each skill is independent, meaning you only install what you need, saving context and routing overhead.
For developers, this is a zero-dependency solution that eliminates lockfile bloat. It enforces strict boundaries, ensuring the model loads the correct skill at the right time. It is ideal for DeepSeek Harness users who want to modularize their workflow, improve agent reliability, and ensure a clear path to verification.
Use Cases
- Detect repetitive work and friction in sessions
- Assess the clarity of the repo's change path
- Verify the integrity of agent assets
Best For
- DeepSeek Harness plugin developers
- Agent system architects
- Teams focused on dependency management
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.