The Agent Skills Directory

[COMMAND_EXECUTION]: The skill workflow includes shell commands for vault discovery, specifically using find, grep, and xargs. These commands use placeholders like <vault-root> and <key terms> which are derived from user input. This pattern presents a potential shell command injection risk if the agent platform does not sanitize these variables before execution. Additionally, the use of xargs without safety flags (like -0) on filenames can lead to unexpected behavior if filenames contain special characters.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process untrusted data from external sources, such as meeting transcripts, brain dumps, and book highlights. The instructions lack boundary markers or explicit 'ignore instructions' delimiters when interpolating this external content into note templates. This creates a vulnerability where malicious instructions embedded in the raw content could influence the agent's behavior during the extraction and linking phases.

obsidian-note-builder