The Agent Skills Directory

[PROMPT_INJECTION]: The skill identified a potential indirect prompt injection surface. It is designed to read, grep, and analyze all notes within an Obsidian vault and then perform edits or consolidations. It lacks explicit instructions to treat the content of those notes as data rather than instructions.
Ingestion points: vault notes accessed via the Read, Grep, and Glob tools.
Boundary markers: Absent. The skill does not instruct the agent to use delimiters or ignore instructions found within the notes during analysis.
Capability inventory: The skill has access to Write and Edit tools, which could be misused if an adversary-controlled note contains instructions the agent obeys.
Sanitization: Absent. There is no instruction to sanitize or validate note content before processing it in a curation workflow.

content-curator