The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes the content of Markdown files from a user-provided vault, creating a surface where malicious instructions within a note could potentially influence the agent's behavior.
Ingestion points: The scripts/vault_tag_inventory.py script reads note content using the read_note function.
Boundary markers: The skill does not employ specific delimiters or warnings to prevent the agent from following instructions embedded within the notes it processes.
Capability inventory: The scripts/vault_tag_inventory.py script possesses filesystem write capabilities through the apply_assignments function using Path.write_text.
Sanitization: No sanitization or filtering of note content is performed to remove potential prompt injection patterns before the agent processes the text for tagging recommendations.
[COMMAND_EXECUTION]: The skill is designed to execute a local Python script, scripts/vault_tag_inventory.py, using the python3 interpreter to perform vault scanning and apply tag updates. This is a functional requirement for the skill's operation.

obsidian-auto-tagger