The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute powerful shell commands including cp -r for recursive copying and rm -rf for recursive deletion of directories. These commands are used to implement the folder merging logic described in Phase 3. While relevant to the skill's purpose of organization, they provide the agent with significant filesystem modification and deletion capabilities.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of existing folder names and file structures within the user's vault.
Ingestion points: Directory names and file counts are ingested via ls and find commands.
Boundary markers: The instructions recommend using quotes for paths with spaces, but no explicit boundary markers or instruction-ignoring delimiters are defined for the data ingestion phase.
Capability inventory: The skill uses ls, find, mkdir, cp, and rm commands across multiple scripts (Phase 1, Phase 3, Phase 4).
Sanitization: The provided bash scripts use direct variable interpolation (e.g., "$VAULT/SourceFolder/") without explicit sanitization or validation of the folder names, which could lead to command injection if a vault contains maliciously crafted directory names.

obsidian-organize