dsh-run2skill
Run the following command in DeepSeek Harness:
dsh plugin install qkycir-123/dsh-run2skill
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install qkycir-123/dsh-run2skill in your terminal to install this plugin, whose open-source repository is located at https://github.com/qkycir-123/dsh-run2skill .
About this plugin
Tired of re-teaching your Agent the same workflows and constraints every time you start a new session? dsh-run2skill is built to stop your hard-earned experience from fading when the chat ends. It automatically distills the explicit corrections, long-term constraints, and standard workflows you teach your Agent in DeepSeek Harness (DSH) into reusable native Skills, so you only need to teach it once in real-world tasks.
Instead of silently making up rules, it follows a strict "draft generation, human review, and confirm to save" pipeline. You can inspect the source evidence and scope of each draft in the plugin settings, and it only writes to the native Skill directory after your explicit approval. As a local-first plugin, it prioritizes privacy by extracting only necessary content for analysis, never sneaking in alternative models or leaking your data.
If you want your Agent to remember your preferences and workflows across sessions without losing absolute control over what gets saved, this plugin is the ideal choice for consolidating your digital assets and boosting AI collaboration efficiency.
Screenshots
Use Cases
- Correcting the Agent's mistakes and enforcing long-term compliance with new rules.
- Setting long-term coding standards or copywriting constraints for specific projects.
- Solidifying complex, multi-step workflows into reusable native skills.
Best For
- Engineers frequently using DSH for complex project development.
- Developers who want the Agent to remember personal preferences and team standards across sessions.
- Privacy-conscious power users who require human review over AI-generated behaviors.
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.