Search indexed text corpora using qmd commands instead of grep.
npx skills add https://github.com/mpsuesser/workspace --skill qmd-cli
Please help me install this Agent Skill. Skill: qmd-cli Source: https://github.com/mpsuesser/workspace/tree/main/dotconfig/pi/agent/skills/__techs__/.unused/qmd-cli Command: npx skills add https://github.com/mpsuesser/workspace --skill qmd-cli
Searching large, indexed text corpora with grep-style techniques is slow, noisy, and error-prone because indexed files (especially JSONL session logs) are difficult to interpret without the right retrieval layer.
Tell the AI: use qmd search in the sessions collection to find the top matches for the phrase apply_edits error, return compact file paths, then inspect the best transcript.
High-intent search queries and answers about installing and using this skill.
Searching indexed text corpora without slow grep commands requires using qmd-based search and retrieval. This approach uses ranked discovery across session logs and documents, avoiding the noise and errors of raw grep on difficult-to-interpret indexed files.
Locating agent session transcripts for debugging tool failures is best done by searching the sessions collection with qmd. You can inspect the best transcript with qmd get, and recover exact raw details using original_session plus session-view tools when needed.
You should use qmd search for exact clues, qmd query for conceptual or weak lexical matches, and qmd vsearch solely for pure semantic similarity. Choosing the correct command based on clue strength ensures efficient and accurate session discovery.
Confirming which collections and contexts are indexed before searching is done using qmd status and discovery commands. This preliminary step verifies the available indexed text corpora so you can target your search and retrieval accurately.
You can use safe shell filtering with tools like awk, sed, sort, and uniq on indexed JSONL session logs. However, you must not directly grep indexed files for meaning, relying instead on qmd for the actual search and retrieval layer.
You need npm to install the qmd package globally and python3 for associated script execution. These dependencies provide the environment necessary to run the qmd-based search and retrieval workflow for inspecting session transcripts.
w-winterSearch indexed text corpora with semantic ranking and structured queries.
mcarcasoRetrieves exact chat messages from a database via SQL queries with structured metadata.
chatermAnalyze server logs with grep, awk, sed, sort, and uniq.
omeraltnMap system workflows into build-ready workflow tree specifications with step-level actions and test cases.
