dsh-continual-harness
Continual self-refinement loop: persistent memory, periodic review-and-refine, cross-session shared knowledge, and automatic rollback through a model-callable harness_refine tool.
持续自我优化闭环：持久记忆、周期评审与自动优化、跨会话共享知识、失败自动回滚，由模型可调用的 harness_refine 工具驱动。
How to install
dsh plugin add dsh-continual-harness About
dsh-continual-harness English | 中文 A **continual self-refinement plugin** for DeepSeek Harness: one plugin gives the agent a closed loop of *persistent memory + periodic review-and-refine + cross-session shared knowledge + automatic rollback on failure* (plan → validate → apply → rollback), implemented through dsh's plugin mechanisms (session events, agent-scoped events, pre-step waterfall, tools service). The design is inspired by the open-source prime-agent from Prime Intellect, a self-improving coding harness. One plugin is enough There is no need to split into multiple packages: this plugi…
Recommendation signals
Meta
- License
- MIT
- Language
- TypeScript
- GitHub stars
- 9
- mo. downloads
- 1.7K
- Last push
- 2026-09-24
- Created
- 2026-08-17
Basic safety check
- Findings
- None
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
- Topics
- agent-memory, ai-agent, continual-learning, deepseek, deepseek-harness, dsh, dsh-plugin, self-evolution, self-improving-agents
Related plugins
langfuse
langfuse/langfuse
🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
claude-code-templates
davila7/claude-code-templates
CLI tool for configuring and monitoring Claude Code
semiotic
nteract/semiotic
React data visualization library for streaming, networks, and AI-assisted development
dsh-desktop
dataelement/dsh-desktop
DSHDesktop：DeepSeek Harness Desktop / DeepSeek Harness 桌面版