The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard shell commands including cat, find, ls, grep, and shell redirection to read and write files within the local filesystem. These operations are limited to the path specified in the OBSIDIAN_VAULT_PATH environment variable or the default Documents directory.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. It reads content from user-controlled files (markdown notes) and presents it to the agent without boundary markers or sanitization. If a note contains malicious instructions, they could potentially influence the agent's behavior during subsequent interactions.
Ingestion points: SKILL.md (Reading note content via cat and grep commands).
Boundary markers: Absent.
Capability inventory: File system read/write (cat, find, ls, grep, redirection).
Sanitization: Absent.

obsidian