The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like find and executes Python scripts to audit the file system, calculate directory sizes, and map the vault structure.
[DYNAMIC_EXECUTION]: The workflow involves generating and running Python scripts at runtime to perform tree-walking and file analysis tasks.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill processes the entire directory structure and file content of a user-provided Obsidian vault.
Boundary markers: None explicitly defined in the instructions to prevent the agent from following instructions embedded within the notes it is auditing.
Capability inventory: The skill has the ability to move files, delete directories, and modify (patch) markdown files in the ~/.hermes/skills/ directory.
Sanitization: There is no mention of sanitizing file paths or note content before they are used to update (patch) the agent's own skill files, which could lead to accidental or malicious modification of agent behavior.

vault-organization