Ingest raw sources and update wiki summaries, index, and logs.
npx skills add https://github.com/skytiger6724/qwen-skills --skill karpathy-wiki
Please help me install this Agent Skill. Skill: karpathy-wiki Source: https://github.com/skytiger6724/qwen-skills/tree/main/karpathy-wiki Command: npx skills add https://github.com/skytiger6724/qwen-skills --skill karpathy-wiki
Keeps a persistent, compounding markdown knowledge base so that raw research is ingested just once and every subsequent query builds on an evolving Karpathy-style wiki instead of redeveloping context from scratch.
Ask the agent to ingest new raw sources into the Karpathy Wiki so the knowledge base stays up to date.
High-intent search queries and answers about installing and using this skill.
A Karpathy-style knowledge wiki structures markdown directories into raw, core knowledge, and workspace sections. It uses a decision tree to coordinate ingestion, queries, and lint checks so raw research is processed once and compounds over time.
To ingest raw sources efficiently, apply the wiki decision tree to process new materials into concept and entity summaries. The workflow automatically updates the index and log records to capture new material while maintaining traceable provenance.
Yes, you can automate health checks across wiki directories. The system provides session-start drift checks and stop hooks that auto-detect new raw content, prompting ingestion when necessary while reporting lint-style health status.
The best way to structure a knowledge wiki is to define raw, core knowledge, and workspace directories. This structured directory layout allows ingestion, queries, linting, and initialization operations to execute consistently across the knowledge base.
Yes, managing this wiki requires shell-level access to the wiki raw folders, the ingest query lint decision tree, and the logging conventions documented in the operations references to execute maintenance commands properly.
Astro-HanCompile raw source materials into structured markdown wiki pages.
weipingDeduplicate concept stubs, assign tiers, and generate a tiered intellectual map.
plurigridIngest Bluesky AT Protocol data into structured schemas and DB-ready artifacts.
jellydnOrganize project learnings and issue notes into a centralized knowledge base via qmd MCP server.
