The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests data from markdown files in the specified vault path using search tools. This creates an indirect prompt injection surface where instructions hidden in notes could be interpreted as commands by the agent.
Ingestion points: Reads file content via grep in /mnt/d/Obsidian Vault/AI Research/.
Boundary markers: Absent; the skill does not use specific markers to delineate untrusted file content from system instructions.
Capability inventory: File search and read capabilities via find and grep in SKILL.md.
Sanitization: None; the content of the markdown files is not validated or sanitized before being processed.
[COMMAND_EXECUTION]: Executes standard shell commands such as find and grep to interact with the local filesystem. These operations are intended for the skill's primary purpose and are limited to the hardcoded vault path.

obsidian-vault