qmd
Warn
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the 'qmd' package globally via 'bun install -g https://github.com/tobi/qmd'. This source is a personal GitHub repository that does not belong to a well-known service or a trusted organization.
- [COMMAND_EXECUTION]: The skill instructions provide the agent with several shell commands to manage software installation, file indexing, and data retrieval, including 'bun install', 'qmd search', and 'qmd get'.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data from local markdown files. * Ingestion points: File content retrieved through 'qmd get' and 'qmd search --full'. * Boundary markers: Absent; there are no instructions to use delimiters or to disregard instructions found within the documents. * Capability inventory: The skill utilizes shell command execution and local file system access. * Sanitization: Absent; the skill does not specify any validation or filtering of the content retrieved from markdown files before it is processed by the AI.
Audit Metadata