The Agent Skills Directory

[PROMPT_INJECTION]: The skill searches and reads Markdown files from a local directory (/mnt/d/Obsidian Vault/AI Research/). This introduces a surface for indirect prompt injection, where notes could contain instructions that influence the agent when they are retrieved.\n
Ingestion points: find and grep operations documented in SKILL.md.\n
Boundary markers: None specified to differentiate between data and instructions.\n
Capability inventory: Execution of search commands (find, grep) and note creation.\n
Sanitization: No validation or filtering of note content is mentioned.\n- [COMMAND_EXECUTION]: The skill instructs the agent to use shell commands like find and grep with user-provided keywords. This pattern can lead to command injection if the agent does not properly escape the keywords before passing them to the shell.

obsidian-vault