empty-fort-strategy
Run the following command in DeepSeek Harness:
dsh plugin install HuaJi2077/empty-fort-strategy
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install HuaJi2077/empty-fort-strategy in DeepSeek Harness to install this plugin; the full source code is available at https://github.com/HuaJi2077/empty-fort-strategy
About this plugin
Empty Fort Strategy is a purely for-fun DeepSeek Harness plugin inspired by the classic Three Kingdoms tale of opening the city gates, burning incense, and playing the guqin to lull the enemy into thinking the city is empty. Its purpose is simple and playful: when you have surplus Token to spend, switch to this mode and the model will dutifully think, invoke a handful of tools, and run through a full multi-round routine, only to close with a single line break and no answer at all, as if to say: there is indeed nothing of use here.
The core capability is a configurable Empty-Fort performance pipeline. You can tune thinkingRounds (number of thinking rounds), toolCalls (number of random tool invocations drawn from the bundled fake-tool pool), and outputFinal (whether the model produces a real result). By default it runs through roughly ten iterations that look like a complete reasoning and tool-use cycle, then stops just before delivering anything concrete, emitting only a blank line. It does not actually answer your question and does not touch any external state, making it a harmless little reasoning observation or simply a bit of fun.
Who is it for? Developers who want to observe how the model handles multi-round reasoning and tool-calling without committing to a real use case; people who want to add a touch of levity to a team chat and give their extra Token spend a story; and simply the curious, who want to see what it looks like when a model is busy for a while and then deliberately does nothing. It does not solve any serious problem. It solves exactly one thing: letting surplus Token be spent, with a flourish, on an Empty-Fort strategy.
Screenshots
Use Cases
- Spend surplus token budget
- Observe multi round reasoning and tool calling
- Add a light moment to team collaboration
Best For
- Developers exploring reasoning flows without committing to a real output
- People who want extra token spend to feel playful
- Curious users who enjoy watching a busy but empty final result
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.