The Agent Skills Directory

[PROMPT_INJECTION]: The skill reads data from local markdown files. While this creates a surface for indirect prompt injection if those files contain malicious instructions, the skill uses grep to specifically target frontmatter keys, which significantly limits the risk of processing arbitrary content.
[COMMAND_EXECUTION]: Shell commands including grep, ls, and basename are used to process local files. These are limited to the user's vault directories and are used for their intended purpose of data retrieval.
[EXTERNAL_DOWNLOADS]: The skill performs a network request to 127.0.0.1:8090. Since this is the local loopback address, it represents interaction with a local task management service rather than an external or untrusted network source.

query