The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates by reading and modifying user-controlled markdown files, creating a surface area for indirect prompt injection attacks where malicious instructions in document content could influence agent behavior. -- Ingestion points: The skill utilizes Read, Glob, and Grep tools to process text content from markdown files within the vault workspace (SKILL.md). -- Boundary markers: The instructions do not define clear delimiters or specific prompts to ignore potential commands embedded within the documents being validated or repaired. -- Capability inventory: The skill possesses the Edit capability to modify files, which could be exploited if an injected instruction successfully overrides the agent's intent during a link-fixing operation (SKILL.md). -- Sanitization: There is no evidence of content sanitization, filtering, or validation logic applied to the ingested text content before the agent acts upon it.

obsidian-links