qmd Search
PassAudited by ClawScan on May 1, 2026.
Overview
This is a coherent local search helper, but users should verify the external qmd CLI and avoid indexing private files they do not want surfaced to the agent.
This skill appears benign and purpose-aligned for local search. Before using it, make sure the qmd CLI you install is trustworthy, understand its local model downloads, and only add collections that you are comfortable having searched and quoted by the agent.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Users need to trust and manage the qmd installation themselves; this skill does not provide provenance for the external CLI or model download source.
The instructions rely on the external `qmd` CLI and local model downloads, but the metadata does not declare the binary source or installation path.
Source: unknown; Homepage: none; Required binaries (all must exist): none; Install specifications: No install spec — this is an instruction-only skill.
Install qmd only from a trusted source, verify the command on your system, and understand where its models are downloaded from before using embedding or reranking features.
Private local documents or code in indexed collections could be surfaced to the agent during answers.
The skill can index local folders and retrieve snippets or full files into the agent's context, which is expected for search but may include private notes, code, or configuration content.
Context gathering — pull relevant snippets before answering questions qmd collection add /path/to/folder --name myproject --mask "*.md,*.py" qmd get myproject/README.md
Use narrowly scoped collections, choose masks carefully, avoid indexing secrets or sensitive personal files, and review results before relying on retrieved context.
