Capture lessons into Markdown training notes with Director confirmation.
npx skills add https://github.com/swishstar/vwpm-assistant --skill remember-swishstar
Please help me install this Agent Skill. Skill: remember Source: https://github.com/swishstar/vwpm-assistant/tree/main/.cursor/skills/remember Command: npx skills add https://github.com/swishstar/vwpm-assistant --skill remember-swishstar
Convert fleeting corrections, preferences, decisions, and discovered patterns into concise, machine-readable training documents so future AI behavior and team processes consistently reflect learned lessons.
When the Director says "remember this" draft a Training Note summarizing the lesson, present it for confirmation, and then save it as training/YYYY-MM-DD-[slug].md.
High-intent search queries and answers about installing and using this skill.
To capture meeting corrections into structured training notes, draft a summary of the lesson immediately after it occurs and save it as a single Markdown file in the training directory. This ensures fleeting lessons become machine-readable documents for future reference.
Yes, the lesson capture process requires Director confirmation before saving files. Drafts of corrections, preferences, or decisions must be presented for approval to ensure accuracy before being stored as plain Markdown training notes.
To organize workflow changes into a machine-readable knowledge base, capture each change as an individual Markdown training note with metadata like date and type. You can later compress these single-lesson files into a centralized knowledge base.
No, you cannot save multiple lessons in a single training note file. The system enforces a one lesson per file rule to keep the training corpus clean and ensure each correction or preference remains a distinct, actionable document.
0-CYBERDYNE-SYSTEMS-0Configure FFT_nano channels, triggers, integrations, and routing behavior.
swishstarExtract lessons from training documents and update KNOWLEDGE_BASE.md.
Galaxies-devConvert audio and video recordings into multilingual transcripts with timestamps.
xsw632Converts natural language hardware descriptions into Chime-compliant JSON specifications with validation and workspace setup.
