The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8) due to its core design of reading and processing external data from the Obsidian vault.
Ingestion points: The skill relies on reading the VAULT.md file and individual notes using the mcp__obsidian__view tool to determine organizational rules and content placement.
Boundary markers: No explicit boundary markers or delimiters are defined to isolate untrusted note content from the agent's internal instructions.
Capability inventory: The skill utilizes several powerful tools including mcp__obsidian__vault (list, create, move), mcp__obsidian__edit (edit, append), and mcp__obsidian__graph (statistics, neighbors, forwardlinks, traverse) which allow the agent to modify the file system and connectivity structure.
Sanitization: The skill does not implement sanitization or validation of the data read from the vault before using it to generate updates or index entries.

obsidian-vault