Transform Markdown documents into CtxFST-formatted semantic chunks with frontmatter metadata.
npx skills add https://github.com/ctxfst/skill-chunk-md --skill skill-chunk-md
Please help me install this Agent Skill. Skill: skill-chunk-md Source: https://github.com/ctxfst/skill-chunk-md/tree/main Command: npx skills add https://github.com/ctxfst/skill-chunk-md --skill skill-chunk-md
Long-form Markdown docs often become hard to navigate and retrieve specific information from, especially when chunk boundaries are unclear and frontmatter metadata is missing.
Provide a Markdown document and the skill will return a CtxFST-formatted version with <Chunk> wrappers and frontmatter.
High-intent search queries and answers about installing and using this skill.
Chunk Markdown documents for semantic retrieval by wrapping text in <Chunk> tags and injecting YAML frontmatter. This preserves meaning and enables precise indexing for pipelines like LanceDB, LightRAG, and HippoRAG.
Semantic chunking splits long-form Markdown into self-contained units using <Chunk> tags. It prevents meaning loss across boundaries and adds frontmatter metadata, ensuring reliable retrieval and reranking for RAG applications.
Yes, LanceDB works seamlessly with Markdown chunks that include YAML frontmatter. The consistent schema containing ids, tags, context, and optional RAG fields ensures reliable indexing and retrieval.
Add YAML frontmatter to Markdown chunks by defining a schema with ids, tags, context, and optional RAG fields. This metadata standardizes chunk boundaries and supports scalable retrieval architectures.
The best way to structure Markdown chunk boundaries is using <Chunk> tags to create self-contained segments. Enforcing a consistent chunk structure with frontmatter metadata ensures reliable retrieval and reranking.
Yes, you can use Markdown chunking with LightRAG and HippoRAG pipelines. The skill formats chunks with consistent frontmatter schemas and <Chunk> tags to support scalable RAG architectures.
KaiserWhoLearnsConvert long text into a single MP3 audio file using Google's gTTS.
ricardorocheStandardizes chunking, vector database integration, embedding management, and retrieval reranking for RAG pipelines.
manifoldlogicConvert code chunks into numerical vectors for semantic codebase search.
yehezkieldioIdentify optimal chunk boundaries for retrieval accuracy across markdown, code, PDF, and plain text documents.
