Search indexed text corpora with semantic ranking and structured queries.
npx skills add https://github.com/w-winter/dot314 --skill text-search
Please help me install this Agent Skill. Skill: text-search Source: https://github.com/w-winter/dot314/tree/main/skills/text-search Command: npx skills add https://github.com/w-winter/dot314 --skill text-search
This skill enables fast, accurate search over indexed text corpora such as session logs, documents, and logs, replacing slow grep-based workflows with semantic ranking and relevance.
Run a basic query like: qmd search "OAuth error" to locate relevant items in the sessions collection, then refine with qmd vsearch for semantic similarity and qmd get to view results.
High-intent search queries and answers about installing and using this skill.
To search indexed text corpora with semantic ranking, you can use qmd-backed indexing which replaces manual grep workflows by providing BM25 keyword search, vector similarity retrieval, and hybrid ranking across sessions, documents, and logs.
The best way to locate specific conversations in session logs is running a keyword query like qmd search "OAuth error" to find matches in the sessions collection, then refining the results using qmd vsearch for semantic similarity.
Yes, you can perform hybrid retrieval over local documents by combining BM25 keyword search with vector similarity. This approach provides ranked results across indexed text collections to quickly locate specific documents or events.
Semantic search works with local collections by using qmd-backed indexing to scan session logs and documents. It supports commands like qmd vsearch for vector similarity and qmd get to retrieve and view the matched results.
You use semantic similarity for searching indexed text because it provides relevance ranking and vector-based retrieval, which offers faster and more accurate results than slow manual grep workflows when locating events across large archives.
A limitation of using qmd commands for searching indexed text is that your sessions, documents, and logs must be properly indexed beforehand to enable BM25 and vector search retrieval across the local collections.
cloudlinessSearch local markdown notes and transcripts with hybrid BM25 and vector retrieval.
v36372Search the alpha-research vault using QMD commands.
mpsuesserSearch indexed text corpora using qmd commands instead of grep.
tiankong0101-byteRetrieve relevant stored content using hybrid semantic and keyword search.
