src/simba/skills/qmd/SKILL.md
Local hybrid search for markdown notes and docs. Use BEFORE reading files to save tokens - search first, read only what's relevant. Provides 90% token savings on exploration tasks.
npx skillsauth add mahmoudimus/simba qmdInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Local search engine for Markdown notes, docs, and knowledge bases. Use this to find relevant files BEFORE reading them to dramatically reduce token usage.
ALWAYS prefer qmd search over reading files directly when:
Token savings: Instead of reading 5 files (5000+ tokens), search first (50 tokens) and read only the relevant sections (200 tokens).
qmd search "query" - Fast BM25 keyword search. Use this first, it's instant.qmd vsearch "query" - Semantic similarity. Use only if keyword search fails.qmd query "query" - Hybrid + reranking. Avoid unless user explicitly requests highest quality.# Fast keyword search (DEFAULT - use this first)
qmd search "authentication" -n 10
# Search specific collection
qmd search "api routes" -c my-project
# Get file paths only (for deciding what to read)
qmd search "database schema" --files
# JSON output for parsing
qmd search "error handling" --json
# Semantic search (slower, use as fallback)
qmd vsearch "how does the login flow work"
# Get specific file
qmd get "path/to/file.md"
# Get by document ID from search results
qmd get "#docid"
# Get multiple files
qmd multi-get "docs/*.md" --json
Bad (token expensive):
1. Read src/auth/login.ts (800 tokens)
2. Read src/auth/session.ts (600 tokens)
3. Read src/auth/middleware.ts (500 tokens)
4. Read src/auth/types.ts (400 tokens)
-> Found answer in middleware.ts
Total: 2300 tokens
Good (token efficient):
1. qmd search "authentication middleware" --files (response: 50 tokens)
-> Returns: src/auth/middleware.ts:45-62
2. Read only the relevant section (150 tokens)
Total: 200 tokens (90% savings!)
qmd status to see indexed collectionsqmd status # Check index health and collections
qmd update # Re-index changed files (fast)
qmd embed # Update vector embeddings (slower)
development
Index the current project for optimized search with QMD semantic search and fast file suggestions. Run this when entering a new codebase or after significant changes. Saves 60-80% tokens on exploration tasks.
tools
Show token economics comparing usage with turbo-search vs without. Demonstrates actual savings from search-first approach.
development
Enforce Simba's Codex lifecycle routine for coding tasks. Use when starting or finishing implementation work in a Simba-enabled repo to run `simba codex-status` at start, `simba codex-extract` when extraction is pending, and `simba codex-finalize` before final handoff.
tools
Save the current work session to persistent memory for future context. Summarizes accomplishments, tracks files modified, and stores learnings for cross-session continuity.