The Agent Skills Directory

[COMMAND_EXECUTION]: The skill workflow instructs the agent to execute various shell commands including find, grep, sed, comm, and xargs to perform vault analysis and identify orphaned files (SKILL.md, Step 0 and Step 4).
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its processing of user-controlled note content.
Ingestion points: The agent is instructed to read and parse all .md files within the vault root to extract wikilink patterns and identify relationships.
Boundary markers: The instructions do not define any boundary markers or provide guidelines to the agent to ignore or escape potentially malicious instructions embedded within the notes it parses.
Capability inventory: The agent has filesystem read/write access and the ability to execute shell commands to search for files, extract text patterns, and create new notes.
Sanitization: There is no requirement for sanitizing extracted note content or filenames, which could potentially contain shell metacharacters and lead to unintended command execution if improperly handled by the agent.

obsidian-links