Overview
@lizhecome/dsh-prompt-optimizer
LLM-backed prompt optimization bundle for DeepSeek Harness
README / EN
Package documentation
Registry summary
LLM-backed prompt optimization bundle for DeepSeek Harness
dsh.pub verifies the pinned bundle contract, runtime facts, and distribution semantics. The complete README remains in the source repository.
Read the full README on GitHubLIMITATIONS
Known limitations
- Multimodal and mixed-block direct-user messages pass through unchanged. - Prompt optimization is semantic model output, so it can still distort intent; use `append` when auditability matters. - The auxiliary model exchange is not stored as a separate model event. Button results are stored in the durable command lifecycle, while automatically delivered output is stored in the main turn.