dsh-deworm
Run the following command in DeepSeek Harness:
dsh plugin install uckkk/dsh-deworm
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install uckkk/dsh-deworm in DeepSeek Harness to install this plugin; the source repository is available at https://github.com/uckkk/dsh-deworm .
About this plugin
In many agent workflows, certain decision points demand a fast, deterministic piece of advice, yet calling out to an external API adds both latency and nondeterminism. dsh-deworm was built to fill exactly that gap by packaging a lightweight, rule-based recommendation step into a standard workflow tool.
The plugin exposes a single tool called deworm (the Deworm Plan): it receives a set of parameters and returns a clear, definitive suggestion without hedging. The entire implementation is pure Node with zero network dependencies, so it runs reliably inside offline sandboxes and introduces no extra I/O overhead.
It is a good fit for developers who want a deterministic recommendation node in their pipeline without pulling in third-party services or network calls, especially when latency and reproducibility are top priorities.
Use Cases
- Insert a deterministic recommendation step into an agent workflow without relying on external APIs
- Run lightweight decision logic inside a fully offline sandbox with zero network I/O
- Return a clear, definitive conclusion based on parameters in an automated pipeline that disallows hedging
Best For
- Developers who want a deterministic decision node in their workflow without adding third-party services
- Automation engineers sensitive to latency and requiring reproducible output
- Independent developers building agent workflows in offline or restricted environments
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.