The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to interact with the file system. It executes git rev-parse, find, grep, and cat based on paths stored in a registry file. While these are used for the stated purpose of note searching, they involve direct execution of shell utilities on local data.
[EXTERNAL_DOWNLOADS]: The skill facilitates data ingestion from external local directories (Obsidian, Logseq, markdown folders). This represents an ingestion point for untrusted data that could contain malicious instructions designed to influence the agent's behavior (Indirect Prompt Injection).
[COMMAND_EXECUTION]: The skill uses $ARGUMENTS in a context that suggests shell interpolation, which could lead to command injection if argument sanitization is not handled by the platform's core execution engine.
[COMMAND_EXECUTION]: Capability Inventory: The skill performs subprocess calls using find, grep, cat, and date across various modes (List, Connect, Search, Context reading). This provides a significant capability tier for an agent interacting with potentially untrusted note files.

vault