Refresh a project's vector index with recent code changes and restart file watchers.
npx skills add https://github.com/dmzoneill/skills --skill knowledge-refresh
Please help me install this Agent Skill. Skill: knowledge-refresh Source: https://github.com/dmzoneill/skills/tree/main/knowledge-refresh Command: npx skills add https://github.com/dmzoneill/skills --skill knowledge-refresh
This Skill ensures your project's knowledge base, including its vector index, is current with recent code changes and architectural updates, preventing outdated information from hindering development.
full_rescan=true to ensure the AI has a completely up-to-date understanding of the project's structure and code.Refresh the knowledge index for the current project, restarting the file watcher.
High-intent search queries and answers about installing and using this skill.
A full knowledge rescan completely re-evaluates the entire project structure and architecture to update the vector index. You should trigger this after a significant refactoring effort to ensure the AI has a completely up-to-date understanding of the codebase.
File watchers keep project knowledge up-to-date by monitoring the codebase for modifications and triggering real-time updates to the vector index. Managing these watchers ensures that recent code changes are immediately reflected in the AI's knowledge base.
You need to refresh your project's knowledge base when recent code changes or architectural updates occur. This prevents outdated information from hindering development and maintains an accurate AI understanding of the software project.
Yes, you can restart file watchers to ensure real-time code updates are reflected in the knowledge base. This management process restarts the monitoring scripts to capture recent modifications and update the vector index accurately.
az9713Initialize a project with dark code prevention directories, CLAUDE.md section, and baseline audit.
peterlianpiCreate document embeddings and index them in a vector store for retrieval-augmented generation.
ourinesLaunch background agent sessions for large coding tasks via git worktrees and tmux.
ialamehIndex codebases into LanceDB vector stores and retrieve relevant snippets via natural language queries.