Extract learned skills and decision heuristics from ongoing conversations.
npx skills add https://github.com/Rheinmir/skills-kit --skill learner-rheinmir
Please help me install this Agent Skill. Skill: learner Source: https://github.com/Rheinmir/skills-kit/tree/main/skills/learner Command: npx skills add https://github.com/Rheinmir/skills-kit --skill learner-rheinmir
This skill enables extracting a learned skill from the current conversation, turning informal patterns into reusable insights, and guiding future decisions.
Tell the system to capture a newly discovered skill as a reusable pattern.
High-intent search queries and answers about installing and using this skill.
To extract domain knowledge from a conversation, this skill identifies learned patterns during extended dialogues and distills them into reusable insights. It separates stable workflows from evolving expertise to guide future decisions.
The best way to capture informal patterns as reusable insights is by triggering the skill during an active discussion. It stabilizes extraction steps into a defined workflow while preserving the acquired domain expertise in a separate metadata section.
You can update extracted expertise without destabilizing the workflow because the architecture separates the two. Improvement cycles modify the expertise section while maintaining the stable workflow defined by frontmatter metadata.
Defining a learned skill requires frontmatter-defined metadata with explicit expertise and workflow sections. This structure ensures the extracted domain knowledge and decision heuristics are properly separated from the operational steps.
You should use skill extraction during extended dialogues where complex domain knowledge and decision heuristics emerge organically. It captures evolving expertise dynamically, unlike static manual documentation that quickly becomes outdated.
levi-qiaoExtract and integrate markdown, PDF, and text files into a structured wiki knowledge base.
fmflurryConverts recurring user corrections into reusable rules for future decisions.
tttolStore structured memories with frontmatter metadata for cross-session recall.
ZeliperDelegate coding tasks to a coder-agent via the Task tool.
