The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script scripts/obsidian-helper.py using the uv run command to interact with the file system, calculate paths, and manage Obsidian vault content.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the user's Obsidian vault.
Ingestion points: The script scripts/obsidian-helper.py reads the content of existing markdown files from the user's Obsidian vaults in the read, recent, and context actions.
Boundary markers: Content read from files is delimited in the output by markers such as --- RECENT --- and --- TODAY ---, but there are no instructions to the agent to disregard any instructions found within those files.
Capability inventory: The skill has the ability to write to files (via the platform's Edit tool) and execute shell commands (via uv run).
Sanitization: The skill does not perform any sanitization or validation of the content read from the markdown files before providing it to the agent's context.

obsidian-diary